Skip to content

Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

docs: fix i18n switch, add v0.3.13/v0.3.14, normalize changelog, update Roadmap (#1845)

Zayn Jarvis0f21cf66f748
diff --git a/docs/.vitepress/config.ts b/docs/.vitepress/config.ts
index 87718eca7..8bcbfb5e4 100644
--- a/docs/.vitepress/config.ts
+++ b/docs/.vitepress/config.ts
@@ -34,13 +34,13 @@ const navLabels = {
guide: 'Guide',
api: 'API Reference',
faq: 'FAQ',
- changelog: 'Changelog'
+ about: 'About'
},
zh: {
guide: '指南',
api: 'API 参考',
faq: '常见问题',
- changelog: '更新日志'
+ about: '关于'
}
}
@@ -100,14 +100,14 @@ const enNav: DefaultTheme.NavItem[] = [
{ text: navLabels.en.guide, link: '/en/getting-started/01-introduction', activeMatch: '/en/(getting-started|concepts|guides|agent-integrations)/' },
{ text: navLabels.en.api, link: '/en/api/01-overview', activeMatch: '/en/api/' },
{ text: navLabels.en.faq, link: '/en/faq/faq', activeMatch: '/en/faq/' },
- { text: navLabels.en.changelog, link: '/en/about/02-changelog', activeMatch: '/en/about/' }
+ { text: navLabels.en.about, link: '/en/about/01-about-us', activeMatch: '/en/about/' }
]
const zhNav: DefaultTheme.NavItem[] = [
{ text: navLabels.zh.guide, link: '/zh/getting-started/01-introduction', activeMatch: '/zh/(getting-started|concepts|guides|agent-integrations)/' },
{ text: navLabels.zh.api, link: '/zh/api/01-overview', activeMatch: '/zh/api/' },
{ text: navLabels.zh.faq, link: '/zh/faq/faq', activeMatch: '/zh/faq/' },
- { text: navLabels.zh.changelog, link: '/zh/about/02-changelog', activeMatch: '/zh/about/' }
+ { text: navLabels.zh.about, link: '/zh/about/01-about-us', activeMatch: '/zh/about/' }
]
function collectAllMdFiles(srcDir: string): { relativePath: string; absPath: string }[] {
@@ -244,16 +244,6 @@ export default defineConfig({
search: {
provider: 'local'
},
- nav: enNav,
- sidebar: {
- '/en/getting-started/': localizedGuideSidebarItems('en'),
- '/en/concepts/': localizedGuideSidebarItems('en'),
- '/en/guides/': localizedGuideSidebarItems('en'),
- '/en/agent-integrations/': localizedGuideSidebarItems('en'),
- '/en/api/': localizedReferenceSidebarItems('en'),
- '/en/about/': localizedAboutSidebarItems('en'),
- '/design/': designSidebar
- },
socialLinks: [
{ icon: 'github', link: `https://github.com/${repo}` }
],
@@ -263,15 +253,27 @@ export default defineConfig({
}
},
locales: {
- root: {
+ en: {
label: 'English',
lang: 'en-US',
- link: '/en/getting-started/01-introduction'
+ link: '/en/',
+ themeConfig: {
+ nav: enNav,
+ sidebar: {
+ '/en/getting-started/': localizedGuideSidebarItems('en'),
+ '/en/concepts/': localizedGuideSidebarItems('en'),
+ '/en/guides/': localizedGuideSidebarItems('en'),
+ '/en/agent-integrations/': localizedGuideSidebarItems('en'),
+ '/en/api/': localizedReferenceSidebarItems('en'),
+ '/en/about/': localizedAboutSidebarItems('en'),
+ '/design/': designSidebar
+ }
+ }
},
zh: {
label: '简体中文',
lang: 'zh-CN',
- link: '/zh/getting-started/01-introduction',
+ link: '/zh/',
title: 'OpenViking',
description: '面向 AI Agent 的开源上下文数据库',
themeConfig: {
diff --git a/docs/en/about/01-about-us.md b/docs/en/about/01-about-us.md
index 70d27ffa3..ef5ba9f61 100644
--- a/docs/en/about/01-about-us.md
+++ b/docs/en/about/01-about-us.md
@@ -1,4 +1,4 @@
-# OpenViking Team and Organization
+# Team & Org
## Project Overview
diff --git a/docs/en/about/02-changelog.md b/docs/en/about/02-changelog.md
index 27f507647..9db633b59 100644
--- a/docs/en/about/02-changelog.md
+++ b/docs/en/about/02-changelog.md
@@ -3,35 +3,56 @@
All notable changes to OpenViking will be documented in this file.
This changelog is automatically generated from [GitHub Releases](https://github.com/volcengine/OpenViking/releases).
+## v0.3.14 (2026-04-30)
+
+### Highlights
+
+- **Observability**: OTLP export now supports custom `headers` for traces, logs, and metrics, enabling direct connection to backends that require extra auth or gRPC metadata.
+- **Upload**: Local directory scans and uploads now respect root and nested `.gitignore` rules, reducing noise from build artifacts and temp files.
+- **Search**: `search` and `find` now accept multiple target URIs for cross-directory and cross-repo retrieval.
+- **Multi-tenant**: OpenClaw plugin clarifies `agent_prefix` as prefix-only; OpenCode memory plugin adds tenant header forwarding.
+- **Admin**: New agent namespace discovery API, CLI command, and docs for listing existing agent namespaces under an account.
+
+### Upgrade Notes
+
+- OTLP backends requiring extra auth can now use `headers` across all three exporter types (gRPC metadata in gRPC mode, HTTP headers in HTTP mode).
+- Local directory uploads will now filter files per `.gitignore` by default — previously imported temp/generated files may be excluded after upgrade.
+- OpenClaw plugin `agent_prefix` is now prefix-only and no longer treated as a full agent identifier; docs migrate `agentId` → `agent_prefix`.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.13...v0.3.14)
+
+## v0.3.13 (2026-04-29)
+
+### Highlights
+
+- **Native MCP endpoint**: `openviking-server` now exposes `/mcp` on the same port as the REST API, reusing API-Key auth and providing 9 tools (`search`, `read`, `list`, `store`, `add_resource`, `grep`, `glob`, `forget`, `health`).
+- **User-level privacy configs**: New `/api/v1/privacy-configs` API and `openviking privacy` CLI for managing sensitive skill settings with version history and rollback.
+- **Observability upgrade**: Unified `server.observability` config enables Prometheus `/metrics` and OpenTelemetry exporters for metrics, traces, and logs.
+- **Retrieval tuning**: New `embedding.text_source`, `embedding.max_input_tokens`, `retrieval.hotness_alpha`, and `retrieval.score_propagation_alpha` controls.
+- **API semantics**: Empty search queries rejected early; stricter `viking://` URI validation; standard error envelopes for processing/zip/HTTP errors.
+- **Docker experience**: Persistent state consolidated under `/app/.openviking`; missing `ov.conf` returns 503 initialization guide instead of crashing.
+- **Security**: Bot image tool sandboxed from host filesystem; health checks skip identity resolution without credentials; API key hashing is now an explicit separate switch.
+
+### Upgrade Notes
+
+- `encryption.api_key_hashing.enabled` must now be configured explicitly (defaults to `false`). If you relied on implicit hashing, add it to your config.
+- OpenClaw plugin is remote-only (no local subprocess), `agentId` → `agent_prefix`, `recallTokenBudget` → `recallMaxInjectedChars`.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.12...v0.3.13)
+
## v0.3.12 (2026-04-24)
-## What's Changed
-* feat(semantic): add output_language_override to pin summary/overview language
-* docs(bot): sync gateway config example with #1640 security defaults
-* fix(parser): validate feishu config limits
-* fix(code-hosting): recognize SSH repository hosts with userinfo
-* docs(design): sync §4.6 role_id rules with #1643 passthrough behavior
-* feat: 补充百炼记忆库 LoCoMo benchmark 评测脚本
-* docs(contributing): add maintainer routing map
-* Revert "feat: 补充百炼记忆库 LoCoMo benchmark 评测脚本"
-* [fix] Fix/oc2ov test stability
-* fix(server): map AGFS URI errors
-* feat(git): added azure devops support
-* feat: support trusted admin tenant management
-* fix(session): count tool parts in pending tokens
-* fix(user_id): support dot
-* Add VitePress docs site and Pages deployment
-* feat(parse): support larkoffice.com Feishu document URLs
-* feat(ragfs): add s3 key normalization encoding
-* fix: apikey security: API Key 管理重构与安全增强
-* add Copy Markdown button and llms.txt support
-* fix(api_keys): proxy get_user_role in NewAPIKeyManager (trusted-mode 500 regression)
-* Fix docs npm registry for release builds
-* fix: account name security
+### Highlights
-## v0.3.10 (2026-04-23)
+- **New Integrations**: Added Azure DevOps support for Git hosting and larkoffice.com URL parsing for Feishu documents.
+- **Security**: Overhauled API key management with security hardening, fixed account name exposure, and resolved a trusted-mode 500 regression in proxy role lookups.
+- **Documentation**: Launched a VitePress-powered docs site with GitHub Pages deployment, added llms.txt support and a Copy Markdown button.
+- **Bug Fixes**: Corrected Feishu config limit validation, SSH repository host recognition with userinfo, AGFS URI error mapping, and token counting for pending tool parts.
+- **Developer Experience**: Added maintainer routing map to contributing docs and S3 key normalization encoding for RAGFS.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.10...v0.3.12)
-### Overview
+## v0.3.10 (2026-04-23)
### Highlights
@@ -42,18 +63,6 @@ This changelog is automatically generated from [GitHub Releases](https://github.
- Added a SQLite backend for QueueFS with persisted queues, ack support, and stale processing message recovery.
- Added Locomo / VikingBot evaluation preflight checks and result validation.
-### New Feature Usage
-
-#### Use the new VLM providers
-
-#### Create a new resource file
-
-#### Use VikingDB API-key data-plane mode
-
-#### Install and configure the OpenClaw plugin
-
-#### QueueFS SQLite backend
-
### Improvements
- Adjusted the default `recallTokenBudget` and `recallMaxContentChars` to reduce the risk of overlong OpenClaw auto-recall context injection.
@@ -70,854 +79,376 @@ This changelog is automatically generated from [GitHub Releases](https://github.
- Fixed LiteLLM `cache_control` 400 errors for Gemini + tools and added support for OpenAI reasoning model families.
- Fixed S3FS directory mtime stability, Rust native build environment pollution, and SQLite database extension parsing.
-### Docs, Tests, and Security
-
-- Documented VLM providers, Codex OAuth, Kimi/GLM, `write(mode=create)`, `tools.mcp_servers`, `ov_tools_enable`, Feishu thread settings, and VLM timeout.
-- Added tests for resource builds, Context Engine, the OpenClaw plugin, content write, VLM providers, setup wizard, server bootstrap, and security behavior.
-- Added `SECURITY.md` and updated the README files, multilingual docs, and community QR code.
-- Addressed multiple code scanning and runtime security findings.
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.9...v0.3.10)
## v0.3.9 (2026-04-18)
-## What's Changed
-* reorg: remove golang depends
-* Feat/mem opt
-* fix: openai like embedding models fix, no more matryoshka error
-* feat(bot): Add disable OpenViking config for channels.
-* fix(config): point missing-config help messages to openviking.ai docs
-* fix(embedder): initialize async client state in VolcengineSparseEmbedder
-* feat(examples): add Codex memory plugin example
-* feat(openclaw-plugin): add unified ov_import and ov_search
-* feat(bot): add MCP client support (port from HKUDS/nanobot v0.1.5)
-* feat(eval):Readme add qa
-* feat(cli): support for default file/dir ignore config in `ovcli.conf`
-* benchmark: add LoCoMo evaluation for Supermemory
-* fix(embedder): report configured provider in slow-call logs
-* fix(queue): preserve embedding message ids across serialization
-* test(security): add unit tests for network_guard and zip_safe modules
-* fix(semantic): preserve repository hierarchy in overviews
-* fix(tests): align pytest coverage docs with required setup
-* feat: rerank support extra headers
-* fix: reload legacy session rows
-* fix: protect global watch-task control files from non-root access
-* fix(agfs): enable agfs s3 plugin default
-* fix(claude-code-memory-plugin): improve Windows compatibility
-* fix(pdf): resolve bookmark page mapping
-* fix: update observer test to use /models endpoint instead of non-exis…
-* fix(openclaw-plugin): extend default Phase 2 commit wait timeout
-* pref(retrieve): Optimize the search performance of larger directories by skipping redundant target_directories scope
-* Add third_party directory to Dockerfile
-* Fix/openclaw addmsg
-* feat(bot):Heartbeat fix
-* feat: add `openviking-server init` interactive setup wizard for local Ollama model deployment
-* fix(volcengine): update default doubao embedding model
-* feat: add Memory V2 full suite test
-* update new wechat group qr code
-* feat(filesystem): support directory descriptions on mkdir
-* feat(memory): default to memory v2
-* fix: resolve OpenClaw session file lock conflicts in oc2ov tests
-* fix: isolate temp scope by user within an account
-* fix(docker): raise Rust toolchain for ragfs image builds
-* feat(metric): add metric system
-* openclaw refactor: assemble context partitioning (Instruction/Archive/Session/…
-* Fix/memory v2
-* Fix/memory v2
-* reorg: split parser layer to 2-layer: accessor and parser, so that we can reuse more code
-* fix: merge error
+### Highlights
+
+- **Memory**: Shipped Memory V2 as the new default, including a full test suite, session row migration, and a fix for file lock conflicts in concurrent scenarios.
+- **OpenClaw**: Refactored context partitioning into Instruction/Archive/Session layers, unified `ov_import` and `ov_search` in the plugin, and extended Phase 2 commit wait timeout.
+- **Bot & MCP**: Ported MCP client support from HKUDS/nanobot v0.1.5, added per-channel OpenViking config disable, and fixed heartbeat reliability.
+- **Search & Retrieval**: Optimized large-directory search by skipping redundant scope checks, fixed sparse embedder async initialization, and added rerank extra-headers support.
+- **Setup & Onboarding**: Introduced an interactive `openviking-server init` wizard for local Ollama deployment and added a default file/dir ignore config for `ovcli.conf`.
+- **Infrastructure**: Added a metric system, updated the default Doubao embedding model, raised the Rust toolchain for RAGFS Docker builds, and split the parser layer into accessor and parser sublayers.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.8...v0.3.9)
## v0.3.8 (2026-04-15)
### Memory V2 Spotlight
-- Format:
-- Refactor and optimization:
-- What users get from this:
+Memory V2 is now the default memory pipeline, featuring a redesigned format, refactored extraction and dedup flow, and improved long-term memory quality.
### Highlights
-- Memory V2 by default:
-- Local deployment and setup:
-- Plugin and agent ecosystem improvements:
-- Config and deployment improvements:
-- Performance and reliability:
-- Other additions:
+- Memory V2 by default with improved format and extraction pipeline.
+- Local deployment and setup experience enhancements (`openviking-server init`).
+- Plugin and agent ecosystem improvements (Codex, OpenClaw, OpenCode examples).
+- Config and deployment improvements (S3 batch delete toggle, OpenRouter `extra_headers`).
+- Performance and reliability improvements across memory, session, and storage layers.
### Upgrade Notes
- If you frequently upload directories through the CLI, consider setting `upload.ignore_dirs` in `ovcli.conf` to reduce noisy uploads.
-- If you need legacy behavior, you can explicitly set `"memory": { "version": "v1" }` in `ov.conf` to fall back to the v1 memory pipeline.
-- If you previously used `ov init` or `ov doctor`, switch to `openviking-server init` and `openviking-server doctor`.
-- If you use OpenRouter or other OpenAI-compatible rerank/VLM providers, `extra_headers` can now be used to inject required headers.
-- If you run against Alibaba Cloud OSS or other S3-compatible services with batch-delete quirks, consider enabling `storage.agfs.s3.disable_batch_delete`.
-- If you are building agent integrations, review the updated examples under `examples/codex-memory-plugin` and `examples/openclaw-plugin`.
+- Legacy memory v1 can be restored via `"memory": { "version": "v1" }` in `ov.conf`.
+- `ov init` / `ov doctor` → `openviking-server init` / `openviking-server doctor`.
+- OpenRouter/compatible rerank/VLM providers can use `extra_headers` for required headers.
+- S3-compatible services with batch-delete quirks: enable `storage.agfs.s3.disable_batch_delete`.
-### Thanks
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.5...v0.3.8)
## v0.3.5 (2026-04-10)
-## What's Changed
-* fix(memory): define config before v2 memory lock retry settings access
-* fix: 优化测试关键词匹配和移除 Release Approval Gate
-* fix: sanitize internal error details in bot proxy responses
-* feat: add scenario-based API tests
-* fix(security): remove leaked token from settings.py
-* Fix/add resource cover
-* Revert "Fix/add resource cover"
-* fix: litellm embedding dimension adapts
-* ci: optimize runner usage with conditional OS matrix and parallel limit
-* fix(bot):Response language, Multi user memory commit
-* ci: remove lite and full test workflows
-* docs: fix docker deployment
-* docs(openclaw-plugin): add health check tools guide
-* fix(queue): expose re-enqueue counts in queue status
-* feat(s3fs): add disable_batch_delete option for OSS compatibility
-* Fix/add resource cover
-* afterTurn: store messages with actual roles and skip heartbeat messages
-* fix: fall back to prefix filters for volcengine path scope
-* fix(session): auto-create missing sessions on first add
-* Fix/api test issues
-* fix: derive context_type from URI in index_resource
+### Highlights
-## v0.3.4 (2026-04-09)
+- **Storage**: Added a `disable_batch_delete` option to S3FS for OSS compatibility and improved RAGFS path scope fallback to prefix filters.
+- **Session & Memory**: Fixed auto-creation of missing sessions on first message add and resolved a Memory V2 config initialization ordering issue.
+- **Bot**: Fixed multi-user memory commits, response language handling, and ensured `afterTurn` stores messages with correct roles while skipping heartbeat entries.
+- **Security & CI**: Removed a leaked token from settings.py, sanitized internal error details in bot proxy responses, and streamlined CI with a conditional OS matrix.
+- **Developer Experience**: Added scenario-based API tests and exposed re-enqueue counts in queue status for easier debugging.
-# OpenViking v0.3.4
-
-## 版本亮点
-
-- **OpenClaw 插件与评测体验继续完善**:调整 `recallPreferAbstract` 与 `ingestReplyAssist` 的默认值以降低意外行为,[PR #1204](https://github.com/volcengine/OpenViking/pull/1204) [PR #1206](https://github.com/volcengine/OpenViking/pull/1206);新增 OpenClaw eval shell 脚本并修复评测导入问题,[PR #1287](https://github.com/volcengine/OpenViking/pull/1287) [PR #1305](https://github.com/volcengine/OpenViking/pull/1305);同时补齐 autoRecall 搜索范围与查询清洗、截断能力,[PR #1225](https://github.com/volcengine/OpenViking/pull/1225) [PR #1297](https://github.com/volcengine/OpenViking/pull/1297)。
-- **Memory、会话写入与运行时稳定性明显增强**:写接口引入 request-scoped wait 机制,[PR #1212](https://github.com/volcengine/OpenViking/pull/1212);补强 PID lock 回收、召回阈值绕过、孤儿 compressor 引用、async contention 和 memory 语义批处理等问题,[PR #1211](https://github.com/volcengine/OpenViking/pull/1211) [PR #1301](https://github.com/volcengine/OpenViking/pull/1301) [PR #1304](https://github.com/volcengine/OpenViking/pull/1304);并继续优化 memory v2 compressor 锁重试控制,[PR #1275](https://github.com/volcengine/OpenViking/pull/1275)。
-- **安全与网络边界进一步收紧**:HTTP 资源导入补齐私网 SSRF 防护,[PR #1133](https://github.com/volcengine/OpenViking/pull/1133);trusted mode 在无 API key 时被限制为仅允许 localhost,[PR #1279](https://github.com/volcengine/OpenViking/pull/1279);embedding circuit breaker 与日志抑制也变得可配置,[PR #1277](https://github.com/volcengine/OpenViking/pull/1277)。
-- **生态与集成能力继续扩展**:新增 Volcengine Vector DB STS Token 支持,[PR #1268](https://github.com/volcengine/OpenViking/pull/1268);新增 MiniMax-M2.7 与 MiniMax-M2.7-highspeed provider 支持,[PR #1284](https://github.com/volcengine/OpenViking/pull/1284);AST 侧补充 Lua parser,[PR #1286](https://github.com/volcengine/OpenViking/pull/1286);Bot 新增 channel mention 能力,[PR #1272](https://github.com/volcengine/OpenViking/pull/1272)。
-- **发布、Docker 与 CI 链路更稳健**:发布时自动更新 `main` 并增加 Docker Hub push,[PR #1229](https://github.com/volcengine/OpenViking/pull/1229);修复 Docker maturin 路径并将 Gemini optional dependency 纳入镜像,[PR #1295](https://github.com/volcengine/OpenViking/pull/1295) [PR #1254](https://github.com/volcengine/OpenViking/pull/1254);CI 侧优化 API matrix、补充 timeout/SMTP 通知并修复 reusable workflow / action 问题,[PR #1281](https://github.com/volcengine/OpenViking/pull/1281) [PR #1293](https://github.com/volcengine/OpenViking/pull/1293) [PR #1300](https://github.com/volcengine/OpenViking/pull/1300) [PR #1302](https://github.com/volcengine/OpenViking/pull/1302) [PR #1307](https://github.com/volcengine/OpenViking/pull/1307)。
-
-## 升级说明
-
-- OpenClaw 插件默认配置发生调整:`recallPreferAbstract` 与 `ingestReplyAssist` 现在默认均为 `false`。如果你之前依赖默认开启行为,升级后需要显式配置,见 [PR #1204](https://github.com/volcengine/OpenViking/pull/1204) 和 [PR #1206](https://github.com/volcengine/OpenViking/pull/1206)。
-- HTTP 资源导入现在默认更严格地防护私网 SSRF;如果你有合法的内网资源采集场景,升级时建议复核现有接入方式与白名单策略,见 [PR #1133](https://github.com/volcengine/OpenViking/pull/1133)。
-- trusted mode 在未提供 API key 时已限制为仅允许 localhost 访问;如果你此前通过非本地地址使用 trusted mode,需要同步调整部署与鉴权配置,见 [PR #1279](https://github.com/volcengine/OpenViking/pull/1279)。
-- 写接口现在引入 request-scoped wait 机制,相关调用在并发写入下的等待与返回时机会更一致;如果你有依赖旧时序的外部编排逻辑,建议升级后复核行为,见 [PR #1212](https://github.com/volcengine/OpenViking/pull/1212)。
-- server 已支持 `host=none` 以使用双栈网络;如果你在 IPv4/IPv6 混合环境中部署,可考虑调整监听配置,见 [PR #1273](https://github.com/volcengine/OpenViking/pull/1273)。
-- Docker 镜像已纳入 Gemini optional dependency,并修复了 maturin 路径问题;如果你维护自定义镜像或发布流水线,建议同步检查构建脚本,见 [PR #1254](https://github.com/volcengine/OpenViking/pull/1254) 和 [PR #1295](https://github.com/volcengine/OpenViking/pull/1295)。
-
-### OpenClaw 插件、评测与集成
-
-- 默认将 openclaw-plugin 的 `recallPreferAbstract` 设为 `false`,[PR #1204](https://github.com/volcengine/OpenViking/pull/1204) by @wlff123
-- 修复 eval 中的 async import 问题,[PR #1203](https://github.com/volcengine/OpenViking/pull/1203) by @yeshion23333
-- 默认将 openclaw-plugin 的 `ingestReplyAssist` 设为 `false`,[PR #1206](https://github.com/volcengine/OpenViking/pull/1206) by @wlff123
-- 向 OpenAIVLM client 增加 timeout 参数,[PR #1208](https://github.com/volcengine/OpenViking/pull/1208) by @highland0971
-- 防止 redo recovery 中阻塞式 VLM 调用导致启动 hang 住,[PR #1226](https://github.com/volcengine/OpenViking/pull/1226) by @mvanhorn
-- 为插件 autoRecall 增加 `skills` 搜索范围,[PR #1225](https://github.com/volcengine/OpenViking/pull/1225) by @mvanhorn
-- 新增 bot channel mention 能力,[PR #1272](https://github.com/volcengine/OpenViking/pull/1272) by @yeshion23333
-- 新增 OpenClaw eval shell 脚本,[PR #1287](https://github.com/volcengine/OpenViking/pull/1287) by @yeshion23333
-- 新增 `add_message` 的 create time,[PR #1288](https://github.com/volcengine/OpenViking/pull/1288) by @wlff123
-- 对 OpenClaw recall query 做清洗并限制长度,[PR #1297](https://github.com/volcengine/OpenViking/pull/1297) by @qin-ctx
-- 修复 OpenClaw eval 导入到 OpenViking 时默认 user 的问题,[PR #1305](https://github.com/volcengine/OpenViking/pull/1305) by @yeshion23333
-
-### Memory、会话、存储与运行时
-
-- 修复明文文件长度小于 4 字节时 decrypt 抛出 `Ciphertext too short` 的问题,[PR #1163](https://github.com/volcengine/OpenViking/pull/1163) by @yc111233
-- 为短明文加密补充单元测试,[PR #1217](https://github.com/volcengine/OpenViking/pull/1217) by @baojun-zhang
-- 修复 PID lock 回收、召回阈值绕过与 orphaned compressor refs 等问题,[PR #1211](https://github.com/volcengine/OpenViking/pull/1211) by @JasonOA888
-- queuefs 去重 memory semantic parent enqueue,[PR #792](https://github.com/volcengine/OpenViking/pull/792) by @Protocol-zero-0
-- 为 decrypt 短明文场景补充回归测试,[PR #1223](https://github.com/volcengine/OpenViking/pull/1223) by @yc111233
-- 为写接口实现 request-scoped wait 机制,[PR #1212](https://github.com/volcengine/OpenViking/pull/1212) by @zhoujh01
-- 将 agfs 重组为 Rust 实现的 ragfs,[PR #1221](https://github.com/volcengine/OpenViking/pull/1221) by @MaojiaSheng
-- 修复 session `_wait_for_previous_archive_done` 无限挂起问题,[PR #1235](https://github.com/volcengine/OpenViking/pull/1235) by @yc111233
-- 修复 `#1238`、`#1242` 和 `#1232` 涉及的问题,[PR #1243](https://github.com/volcengine/OpenViking/pull/1243) by @MaojiaSheng
-- Memory 侧继续做性能优化,[PR #1159](https://github.com/volcengine/OpenViking/pull/1159) by @chenjw
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.4...v0.3.5)
-## v0.3.3 (2026-04-03)
-
-# OpenViking v0.3.3
+## v0.3.4 (2026-04-09)
-## Highlights
+### Highlights
-- **评测与写入能力继续扩展**:新增 RAG benchmark 评测框架 [PR #825](https://github.com/volcengine/OpenViking/pull/825),补充 OpenClaw 的 LoCoMo eval 脚本与说明 [PR #1152](https://github.com/volcengine/OpenViking/pull/1152),并新增内容写入接口 [PR #1151](https://github.com/volcengine/OpenViking/pull/1151)。
-- **OpenClaw 插件可用性显著增强**:补充架构文档与图示 [PR #1145](https://github.com/volcengine/OpenViking/pull/1145),安装器不再覆盖 `gateway.mode` [PR #1149](https://github.com/volcengine/OpenViking/pull/1149),新增端到端 healthcheck 工具 [PR #1180](https://github.com/volcengine/OpenViking/pull/1180),支持 bypass session patterns [PR #1194](https://github.com/volcengine/OpenViking/pull/1194),并在 OpenViking 故障时避免阻塞 OpenClaw [PR #1158](https://github.com/volcengine/OpenViking/pull/1158)。
-- **测试与 CI 覆盖大幅补强**:OpenClaw 插件新增大规模单测套件 [PR #1144](https://github.com/volcengine/OpenViking/pull/1144),补充 e2e 测试 [PR #1154](https://github.com/volcengine/OpenViking/pull/1154),新增 OpenClaw2OpenViking 集成测试与 CI 流水线 [PR #1168](https://github.com/volcengine/OpenViking/pull/1168)。
-- **会话、解析与导入链路更稳健**:支持创建 session 时指定 `session_id` [PR #1074](https://github.com/volcengine/OpenViking/pull/1074),CLI 聊天端点优先级与 `grep --exclude-uri/-x` 能力得到增强 [PR #1143](https://github.com/volcengine/OpenViking/pull/1143) [PR #1174](https://github.com/volcengine/OpenViking/pull/1174),目录导入 UX / 正确性与扫描 warning 契约也进一步改善 [PR #1197](https://github.com/volcengine/OpenViking/pull/1197) [PR #1199](https://github.com/volcengine/OpenViking/pull/1199)。
-- **稳定性与安全性继续加固**:修复任务 API ownership 泄露问题 [PR #1182](https://github.com/volcengine/OpenViking/pull/1182),统一 stale lock 处理并补充 ownership checks [PR #1171](https://github.com/volcengine/OpenViking/pull/1171),修复 ZIP 乱码 [PR #1173](https://github.com/volcengine/OpenViking/pull/1173)、embedder dimensions 透传 [PR #1183](https://github.com/volcengine/OpenViking/pull/1183)、语义 DAG 增量更新缺失 summary 场景 [PR #1177](https://github.com/volcengine/OpenViking/pull/1177) 等问题。
+- OpenClaw plugin defaults adjusted (`recallPreferAbstract` and `ingestReplyAssist` now `false`); eval scripts and recall query sanitization added.
+- Memory and session runtime stability improved: request-scoped write waits, PID lock recovery, orphan compressor refs, async contention fixes.
+- Security tightened: SSRF protection for HTTP resource imports, localhost-only trusted mode without API key, configurable embedding circuit breaker.
+- Ecosystem expansion: Volcengine Vector DB STS Token, MiniMax-M2.7 provider, Lua parser, bot channel mention.
+- CI/Docker: auto-update `main` on release, Docker Hub push, Gemini optional dependency in image.
-## Upgrade Notes
+### Upgrade Notes
-- OpenClaw 插件安装器不再写入 `gateway.mode`。如果你之前依赖安装流程自动改写该配置,升级后需要改为显式管理,见 [PR #1149](https://github.com/volcengine/OpenViking/pull/1149)。
-- 如果你使用 `--with-bot` 进行安装或启动,失败时现在会直接返回错误码;依赖“失败但继续执行”行为的脚本需要同步调整,见 [PR #1175](https://github.com/volcengine/OpenViking/pull/1175)。
-- 如果你接入 OpenAI Dense Embedder,自定义维度参数现在会正确传入 `embed()`;此前依赖默认维度行为的调用方建议复核配置,见 [PR #1183](https://github.com/volcengine/OpenViking/pull/1183)。
-- `ov status` 现在会展示 embedding 与 rerank 模型使用情况,便于排障与环境核对,见 [PR #1191](https://github.com/volcengine/OpenViking/pull/1191)。
-- 检索侧曾尝试加入基于 tags metadata 的 cross-subtree retrieval [PR #1162](https://github.com/volcengine/OpenViking/pull/1162),但已在本版本窗口内回滚 [PR #1200](https://github.com/volcengine/OpenViking/pull/1200),因此不应将其视为 `v0.3.3` 的最终可用能力。
-- `litellm` 依赖范围更新为 `>=1.0.0,<1.83.1`,升级时建议同步检查锁文件与兼容性,见 [PR #1179](https://github.com/volcengine/OpenViking/pull/1179)。
+- OpenClaw `recallPreferAbstract` and `ingestReplyAssist` now default to `false` — enable explicitly if needed.
+- HTTP resource imports now enforce private-network SSRF protection by default.
+- Trusted mode without API key is restricted to localhost only.
+- Write interface now uses request-scoped wait — review external orchestration timing dependencies.
-## What's Changed
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.3...v0.3.4)
-### Benchmark, Eval, CLI, and Writing
+## v0.3.3 (2026-04-03)
-- 新增 RAG benchmark 系统评测框架,[PR #825](https://github.com/volcengine/OpenViking/pull/825) by @sponge225
-- 优化 CLI chat endpoint 配置优先级,[PR #1143](https://github.com/volcengine/OpenViking/pull/1143) by @ruansheng8
-- 新增 OpenClaw 的 LoCoMo eval 脚本与 README,[PR #1152](https://github.com/volcengine/OpenViking/pull/1152) by @yeshion23333
-- 新增内容写入接口,[PR #1151](https://github.com/volcengine/OpenViking/pull/1151) by @zhoujh01
-- `ov cli grep` 新增 `--exclude-uri` / `-x` 选项,[PR #1174](https://github.com/volcengine/OpenViking/pull/1174) by @heaoxiang-ai
-- 更新 rerank 配置文档,[PR #1138](https://github.com/volcengine/OpenViking/pull/1138) by @ousugo
+### Highlights
-### OpenClaw Plugin, OpenCode Plugin, Bot, and Console
+- RAG benchmark evaluation framework added; OpenClaw LoCoMo eval scripts; content write API.
+- OpenClaw plugin: architecture docs, installer no longer overwrites `gateway.mode`, e2e healthcheck tool, bypass session patterns, fault isolation from OpenViking.
+- Test coverage: OpenClaw plugin unit tests, e2e tests, oc2ov integration tests and CI.
+- Session creation now supports specifying `session_id`; CLI chat endpoint priority and `grep --exclude-uri/-x` enhanced.
+- Security: task API ownership leak fix, unified stale lock handling, ZIP encoding fix, embedder dimension passthrough.
-- 刷新 OpenClaw 插件架构文档并补充图示,[PR #1145](https://github.com/volcengine/OpenViking/pull/1145) by @qin-ctx
-- 修复 OpenClaw 插件安装器不应写入 `gateway.mode`,[PR #1149](https://github.com/volcengine/OpenViking/pull/1149) by @LinQiang391
-- 处理 OpenViking 故障时不再阻塞 OpenClaw,[PR #1158](https://github.com/volcengine/OpenViking/pull/1158) by @wlff123
-- 新增 OpenClaw 插件端到端 healthcheck 工具,[PR #1180](https://github.com/volcengine/OpenViking/pull/1180) by @wlff123
-- 新增 bypass session patterns 支持,[PR #1194](https://github.com/volcengine/OpenViking/pull/1194) by @Mijamind719
-- 修复 OpenCode 插件在当前 main 上恢复 stale commit state 的问题,[PR #1187](https://github.com/volcengine/OpenViking/pull/1187) by @13ernkastel
-- 新增 Bot 单通道(BotChannel)集成与 Werewolf demo,[PR #1196](https://github.com/volcengine/OpenViking/pull/1196) by @yeshion23333
-- console 支持 account user agentid,[PR #1198](https://github.com/volcengine/OpenViking/pull/1198) by @zhoujh01
+### Upgrade Notes
-### Sessions, Retrieval, Parsing, and Resource Import
+- OpenClaw installer no longer writes `gateway.mode` — manage explicitly after upgrade.
+- `--with-bot` failures now return error codes; scripts relying on "fail-but-continue" need adjustment.
+- OpenAI Dense Embedder now correctly passes custom dimension to `embed()`.
+- Cross-subtree retrieval via tags metadata was added then reverted in this release window — not a final capability.
+- `litellm` dependency updated to `>=1.0.0,<1.83.1`.
-- 支持创建 session 时指定 `session_id`,[PR #1074](https://github.com/volcengine/OpenViking/pull/1074) by @likzn
-- 修复 reindex 时 memory `context_type` 的 URI 匹配逻辑,[PR #1155](https://github.com/volcengine/OpenViking/pull/1155) by @deepakdevp
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.2...v0.3.3)
## v0.3.2 (2026-04-01)
-## What's Changed
-* Chore/pr agent ark token billing
-* Fix HTTPX recognition issue with SOCKS5 proxy causing OpenViking crash
-* refactor(model): unify config-driven retry across VLM and embedding
-* fix(bot): import eval session time
-* docs(examples): retire legacy integration examples
-* fix(ci): skip filesystem tests on Windows due to FUSE compatibility
-* docs(docker): use latest image tag in examples
-* upload the new wechat group qrcode
-* Unify test directory in openclaw-plugin
-* docs(guides):Add concise OVPack guide in Chinese and English
-* docs(guides): reorganize observability documentation
-* fix(installer): fall back to official PyPI when mirror lags
-* feat(docker) add vikingbot and console
-* fix(vlm): rollback ResponseAPI to Chat Completions, keep tool calls
-* feat(openclaw-plugin): add session-pattern guard for ingest reply assist
-
-## v0.3.1 (2026-03-31)
+### Highlights
-## What's Changed
-* feat(ast): add PHP tree-sitter support
-* feat(ci): add multi-platform API test support for 5 platforms
-* fix(ci): refresh uv.lock for docker release build
-* fix(openclaw-plugin): simplify install flow and harden helpers
-* fix(openclaw-plugin): preserve existing ov.conf on auto install
-* ci: build docker images natively per arch
-* Feature/memory opt
-* feat(storage): add auto language detection for semantic summary generation
-* feat(prompt) support configurable prompt template directories
-* feat: unify config-driven retry across VLM and embedding
-* feat(bot): import ov eval script
-* Revert "feat: unify config-driven retry across VLM and embedding"
-* fix(storage) Fix parent_uri compatibility with legacy records
-* fix(session): unify archive context abstracts
+- **Docker**: Added VikingBot and Console services to Docker setup; updated examples to use latest image tags.
+- **OpenClaw Plugin**: Added session-pattern guard for ingest reply assist; unified test directory structure.
+- **VLM**: Rolled back ResponseAPI to Chat Completions while preserving tool call support.
+- **Reliability**: Fixed HTTPX SOCKS5 proxy crash; improved PyPI mirror fallback in installer; skipped FUSE-incompatible filesystem tests on Windows.
+- **Docs**: Added OVPack guide in Chinese and English; reorganized observability documentation; retired legacy integration examples.
-## v0.2.14 (2026-03-30)
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.3.1...v0.3.2)
-# OpenViking v0.2.14
+## v0.3.1 (2026-03-31)
-## Highlights
+### Highlights
-- 多租户与身份管理进一步完善。CLI 已支持租户身份默认值与覆盖,文档新增多租户使用指南,memory 也支持仅按 agent 维度隔离的 `agent-only` scope。
-- 解析与导入链路更完整。图片解析新增 OCR 文本提取,目录导入识别 `.cc` 文件,重复标题导致的文件名冲突得到修复,HTTP 上传链路改为更稳妥的 upload id 流程。
-- OpenClaw 插件显著增强。安装器与升级流程统一,默认按最新 Git tag 安装,session API 与 context pipeline 做了统一重构,并补齐了 Windows、compaction、compact result mapping、子进程重拉起等多处兼容性与稳定性问题。
-- Bot 与 Feishu 集成可用性继续提升。修复了 bot proxy 未鉴权问题,改进了 Moonshot 请求兼容性,并升级了 Feishu interactive card 的 markdown 展示体验。
-- 存储与运行时稳定性持续提升。包括 queuefs embedding tracker 加固、vector store 移除 `parent_uri`、Docker doctor 检查对齐,以及更细粒度的 eval token 指标。
+- **Language Support**: Added PHP tree-sitter AST parsing.
+- **Storage**: Introduced auto language detection for semantic summary generation; fixed parent URI compatibility with legacy records.
+- **CI**: Expanded API test coverage to 5 platforms; switched to native per-arch Docker image builds; refreshed uv.lock for release.
+- **Configuration**: Added configurable prompt template directories; unified archive context handling in session management.
+- **OpenClaw Plugin**: Simplified install flow, hardened helpers, and preserved existing `ov.conf` on auto install.
+- **Memory**: Applied memory optimization improvements.
-## Upgrade Notes
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.14...v0.3.1)
-- Bot proxy 接口 `/bot/v1/chat` 与 `/bot/v1/chat/stream` 已补齐鉴权,依赖未鉴权访问的调用方需要同步调整,见 [#996](https://github.com/volcengine/OpenViking/pull/996)。
-- 裸 HTTP 导入本地文件/目录时,推荐按 `temp_upload -> temp_file_id` 的方式接入上传链路,见 [#1012](https://github.com/volcengine/OpenViking/pull/1012)。
-- OpenClaw 插件的 compaction delegation 修复要求 `openclaw >= v2026.3.22`,见 [#1000](https://github.com/volcengine/OpenViking/pull/1000)。
-- OpenClaw 插件安装器现在默认跟随仓库最新 Git tag 安装,如需固定版本可显式指定,见 [#1050](https://github.com/volcengine/OpenViking/pull/1050)。
+## v0.2.14 (2026-03-30)
-## What's Changed
+### Highlights
-### Multi-tenant, Memory, and Identity
+- Multi-tenant identity management: CLI tenant defaults and overrides, `agent-only` memory scope, multi-tenant usage guide.
+- Parsing: image OCR text extraction, `.cc` file recognition, duplicate title filename conflict fix, upload-id based HTTP upload flow.
+- OpenClaw plugin: unified installer/upgrade flow, default latest Git tag install, session API and context pipeline refactoring, Windows/compaction/subprocess compatibility fixes.
+- Bot and Feishu: proxy auth fix, Moonshot compatibility, Feishu interactive card markdown upgrade.
+- Storage: queuefs embedding tracker hardening, vector store `parent_uri` removal, Docker doctor alignment, eval token metrics.
-- 支持 CLI 租户身份默认值与覆盖,[#1019](https://github.com/volcengine/OpenViking/pull/1019) by @zhoujh01
-- 支持仅按 agent 隔离的 agent memory scope,[#954](https://github.com/volcengine/OpenViking/pull/954) by @liberion1994
-- 新增多租户使用指南文档,[#1029](https://github.com/volcengine/OpenViking/pull/1029) by @qin-ctx
-- 在 quickstart 和 basic usage 中补充 `user_key` 与 `root_key` 的区别说明,[#1077](https://github.com/volcengine/OpenViking/pull/1077) by @r266-tech
-- 示例中支持将 recalled memories 标记为 used,[#1079](https://github.com/volcengine/OpenViking/pull/1079) by @0xble
+### Upgrade Notes
-### Parsing and Resource Import
+- Bot proxy endpoints `/bot/v1/chat` and `/bot/v1/chat/stream` now require authentication.
+- HTTP file uploads should use the `temp_upload → temp_file_id` flow.
+- OpenClaw plugin compaction delegation requires `openclaw >= v2026.3.22`.
+- OpenClaw installer now defaults to latest Git tag — specify explicitly to pin versions.
-- 新增图片解析 OCR 文本提取能力,[#942](https://github.com/volcengine/OpenViking/pull/942) by @mvanhorn
-- 修复重复标题导致的合并文件名冲突,[#1005](https://github.com/volcengine/OpenViking/pull/1005) by @deepakdevp
-- 目录导入时识别 `.cc` 文件,[#1008](https://github.com/volcengine/OpenViking/pull/1008) by @qin-ctx
-- 增加对非空 S3 目录 marker 的兼容性,[#997](https://github.com/volcengine/OpenViking/pull/997) by @zhoujh01
-- HTTP 上传链路从临时路径切换为 upload id,[#1012](https://github.com/volcengine/OpenViking/pull/1012) by @qin-ctx
-- 修复从目录上传时临时归档文件丢失 `.zip` 后缀的问题,[#1021](https://github.com/volcengine/OpenViking/pull/1021) by @Shawn-cf-o
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.13...v0.2.14)
-### OpenClaw Plugin and Installer
+## v0.2.13 (2026-03-26)
-- 回滚 duplicate registration guard 的加固改动,[#995](https://github.com/volcengine/OpenViking/pull/995) by @qin-ptr
-- Windows 下 mapped session ID 做安全清洗,[#998](https://github.com/volcengine/OpenViking/pull/998) by @qin-ctx
-- 使用 plugin-sdk exports 实现 compaction delegation,修复 `#833`,要求 `openclaw >= v2026.3.22`,[#1000](https://github.com/volcengine/OpenViking/pull/1000) by @jcp0578
-- 统一 installer 升级流程,[#1020](https://github.com/volcengine/OpenViking/pull/1020) by @LinQiang391
-- 默认按最新 Git tag 安装 OpenClaw 插件,[#1050](https://github.com/volcengine/OpenViking/pull/1050) by @LinQiang391
-- 统一 session APIs,并重构 OpenClaw context pipeline 以提升一致性、可维护性和测试覆盖,[#1040](https://github.com/volcengine/OpenViking/pull/1040) by @wlff123
-- 将 openclaw-plugin 的 `DEFAULT_COMMIT_TOKEN_THRESHOLD` 调整为 `20000`,[#1052](https://github.com/volcengine/OpenViking/pull/1052) by @wlff123
+### Highlights
-## v0.2.13 (2026-03-26)
+- **Testing**: Added comprehensive unit tests for core utilities; improved API test infrastructure with dual-mode CI support.
+- **Platform**: Fixed Windows engine wheel runtime packaging.
+- **VLM**: Scoped LiteLLM thinking parameter to DashScope providers only.
+- **OpenClaw Plugin**: Hardened duplicate registration guard.
+- **Docs**: Added basic usage examples and Chinese documentation.
-## What's Changed
-* test: add comprehensive unit tests for core utilities
-* fix(vlm): scope LiteLLM thinking param to DashScope providers only
-* Api test:improve API test infrastructure with dual-mode CI
-* docs: Add basic usage example and Chinese documentation for examples
-* Fix Windows engine wheel runtime packaging
-* fix(openclaw-plugin): harden duplicate registration guard
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.12...v0.2.13)
## v0.2.12 (2026-03-25)
-## What's Changed
-* Use uv sync --locked in Dockerfile
-* fix(server): handle CancelledError during shutdown paths
-* fix(bot):rollback config
+This patch release stabilizes the server shutdown sequence by properly handling `CancelledError`, rolls back a bot configuration regression, and tightens the Docker build by switching to `uv sync --locked` for reproducible dependency resolution.
-## v0.2.11 (2026-03-25)
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.11...v0.2.12)
-# OpenViking v0.2.11
-
-## 版本亮点
-
-- **模型与检索生态继续扩展**:新增 MiniMax embedding [#624](https://github.com/volcengine/OpenViking/pull/624)、Azure OpenAI embedding/VLM [#808](https://github.com/volcengine/OpenViking/pull/808)、GeminiDenseEmbedder [#751](https://github.com/volcengine/OpenViking/pull/751)、LiteLLM embedding [#853](https://github.com/volcengine/OpenViking/pull/853) 与 rerank [#888](https://github.com/volcengine/OpenViking/pull/888),并补充 OpenAI-compatible rerank [#785](https://github.com/volcengine/OpenViking/pull/785) 与 Tavily 搜索后端 [#788](https://github.com/volcengine/OpenViking/pull/788)。
-- **内容接入链路更完整**:音频资源现在支持 Whisper ASR 解析 [#805](https://github.com/volcengine/OpenViking/pull/805),云文档场景新增飞书/Lark 解析器 [#831](https://github.com/volcengine/OpenViking/pull/831),文件向量化策略变为可配置 [#858](https://github.com/volcengine/OpenViking/pull/858),搜索结果还新增了 provenance 元数据 [#852](https://github.com/volcengine/OpenViking/pull/852)。
-- **服务端运维能力明显补齐**:新增 `ov reindex` [#795](https://github.com/volcengine/OpenViking/pull/795)、`ov doctor` [#851](https://github.com/volcengine/OpenViking/pull/851)、Prometheus exporter [#806](https://github.com/volcengine/OpenViking/pull/806)、内存健康统计 API [#706](https://github.com/volcengine/OpenViking/pull/706)、可信租户头模式 [#868](https://github.com/volcengine/OpenViking/pull/868) 与 Helm Chart [#800](https://github.com/volcengine/OpenViking/pull/800)。
-- **多租户与安全能力增强**:新增多租户文件加密 [#828](https://github.com/volcengine/OpenViking/pull/828) 和文档加密能力 [#893](https://github.com/volcengine/OpenViking/pull/893),修复租户上下文在 observer 与 reindex 流程中的透传问题 [#807](https://github.com/volcengine/OpenViking/pull/807) [#820](https://github.com/volcengine/OpenViking/pull/820),并修复 ZIP Slip 风险 [#879](https://github.com/volcengine/OpenViking/pull/879) 与 trusted auth 模式下 API key 校验缺失 [#924](https://github.com/volcengine/OpenViking/pull/924)。
-- **稳定性与工程体验持续加固**:向量检索 NaN/Inf 分数处理在结果端 [#824](https://github.com/volcengine/OpenViking/pull/824) 和源头 [#882](https://github.com/volcengine/OpenViking/pull/882) 双重兜底,会话异步提交与并发提交问题被修复 [#819](https://github.com/volcengine/OpenViking/pull/819) [#783](https://github.com/volcengine/OpenViking/pull/783) [#900](https://github.com/volcengine/OpenViking/pull/900),Windows stale lock 与 TUI 输入问题持续修复 [#790](https://github.com/volcengine/OpenViking/pull/790) [#798](https://github.com/volcengine/OpenViking/pull/798) [#854](https://github.com/volcengine/OpenViking/pull/854),同时补上了代理兼容 [#957](https://github.com/volcengine/OpenViking/pull/957) 与 API 重试风暴保护 [#772](https://github.com/volcengine/OpenViking/pull/772)。
-
-## 升级提示
-
-- 如果你使用 `litellm` 集成,请关注这一版本中的安全策略调整:先临时硬禁用 [#937](https://github.com/volcengine/OpenViking/pull/937),后恢复为仅允许 `<1.82.6` 的版本范围 [#966](https://github.com/volcengine/OpenViking/pull/966)。建议显式锁定依赖版本。
-- 如果你启用 trusted auth 模式,需要同时配置服务端 API key,相关校验已在 [#924](https://github.com/volcengine/OpenViking/pull/924) 中强制执行。
-- Helm 默认配置已切换为更适合 Volcengine 场景的默认值,并补齐缺失字段 [#822](https://github.com/volcengine/OpenViking/pull/822)。升级 chart 时建议重新审阅 values 配置。
-- Windows 用户建议关注 stale PID / RocksDB LOCK 文件处理增强 [#790](https://github.com/volcengine/OpenViking/pull/790) [#798](https://github.com/volcengine/OpenViking/pull/798) [#854](https://github.com/volcengine/OpenViking/pull/854)。
-
-### 模型、Embedding、Rerank 与检索生态
-
-- 新增 MiniMax embedding provider,支持通过官方 HTTP API 接入 MiniMax 向量模型。[PR #624](https://github.com/volcengine/OpenViking/pull/624)
-- 新增 OpenAI-compatible rerank provider。[PR #785](https://github.com/volcengine/OpenViking/pull/785)
-- 新增 Azure OpenAI 对 embedding 与 VLM 的支持。[PR #808](https://github.com/volcengine/OpenViking/pull/808)
-- 新增 GeminiDenseEmbedder 文本向量模型提供方。[PR #751](https://github.com/volcengine/OpenViking/pull/751)
-- 新增 Tavily 作为可配置的 web search backend。[PR #788](https://github.com/volcengine/OpenViking/pull/788)
-- 修复 LiteLLM 下 `zai/` 模型前缀处理,避免重复拼接 `zhipu` 前缀。[PR #789](https://github.com/volcengine/OpenViking/pull/789)
-- 修复 Gemini embedder 相关问题。[PR #841](https://github.com/volcengine/OpenViking/pull/841)
-- 新增 LiteLLM embedding provider。[PR #853](https://github.com/volcengine/OpenViking/pull/853)
-- 新增默认向量索引名可配置能力。[PR #861](https://github.com/volcengine/OpenViking/pull/861)
-- 新增 LiteLLM rerank provider。[PR #888](https://github.com/volcengine/OpenViking/pull/888)
-- 修复 Ollama embedding provider 的维度解析问题。[PR #915](https://github.com/volcengine/OpenViking/pull/915)
-- 为 Jina 代码模型补充 code-specific task 默认值。[PR #914](https://github.com/volcengine/OpenViking/pull/914)
-- 在 embedder 与 vectordb 维度不匹配时给出警告提示。[PR #930](https://github.com/volcengine/OpenViking/pull/930)
-- 为 Jina embedding provider 增加代码模型维度与更可操作的 422 错误提示。[PR #928](https://github.com/volcengine/OpenViking/pull/928)
-- 搜索结果新增 provenance 元数据,方便追踪召回来源。[PR #852](https://github.com/volcengine/OpenViking/pull/852)
-- 修复 recall limit 逻辑。[PR #821](https://github.com/volcengine/OpenViking/pull/821)
-- 在结果序列化前钳制向量检索中的 `inf`/`nan` 分数,避免 JSON 失败。[PR #824](https://github.com/volcengine/OpenViking/pull/824)
-- 在相似度分数源头钳制 `inf`/`nan`,进一步避免 JSON crash。[PR #882](https://github.com/volcengine/OpenViking/pull/882)
-
-### 解析、导入与内容处理
-
-- 新增基于 Whisper 的音频解析与 ASR 集成。[PR #805](https://github.com/volcengine/OpenViking/pull/805)
-- 修复 PDF 书签目标页为整数索引时的解析问题。[PR #794](https://github.com/volcengine/OpenViking/pull/794)
-- 新增飞书/Lark 云文档解析器。[PR #831](https://github.com/volcengine/OpenViking/pull/831)
-- 修复 MarkdownParser 的字符数限制处理。[PR #826](https://github.com/volcengine/OpenViking/pull/826)
-- 支持可配置的文件向量化策略。[PR #858](https://github.com/volcengine/OpenViking/pull/858)
+## v0.2.11 (2026-03-25)
-## v0.2.10 (2026-03-24)
+### Highlights
-# LiteLLM 安全热修复 Release Note
+- Model ecosystem: MiniMax embedding, Azure OpenAI embedding/VLM, GeminiDenseEmbedder, LiteLLM embedding and rerank, OpenAI-compatible rerank, Tavily search backend.
+- Content pipeline: Whisper ASR for audio, Feishu/Lark document parser, configurable file vectorization strategy, search result provenance metadata.
+- Server ops: `ov reindex`, `ov doctor`, Prometheus exporter, memory health stats API, trusted tenant header mode, Helm Chart.
+- Multi-tenant security: file encryption, document encryption, tenant context passthrough fixes, ZIP Slip fix, trusted auth API key enforcement.
+- Stability: vector score NaN/Inf clamping, async/concurrent session commit fixes, Windows stale lock and TUI fixes, proxy compatibility, API retry storm protection.
-更新时间:2026-03-24
+### Upgrade Notes
-## 背景
+- `litellm` security policy: temporarily disabled, then restored as `<1.82.6` — pin your dependency version explicitly.
+- Trusted auth mode now requires a server-side API key.
+- Helm default values updated for Volcengine — review values config on chart upgrade.
-由于上游依赖 `LiteLLM` 出现公开供应链安全事件,OpenViking 在本次热修复中临时禁用所有 LiteLLM 相关入口,以避免继续安装或运行到受影响依赖。
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.10...v0.2.11)
-## 变更内容
+## v0.2.10 (2026-03-24)
-- 移除根依赖中的 `litellm`
-- 移除根 `uv.lock` 中的 `litellm`
-- 禁用 LiteLLM 相关的 VLM provider 入口
-- 禁用 bot 侧 LiteLLM provider 和图片工具入口
-- 增加 LiteLLM 已禁用的回归测试
+### LiteLLM Security Hotfix
-## 建议操作
+Emergency hotfix due to a supply chain security incident in the upstream `LiteLLM` dependency. All LiteLLM-related entry points are temporarily disabled.
-建议用户立即执行以下动作:
+### Action Required
-1. 检查运行环境中是否安装 `litellm`
-2. 卸载可疑版本并重建虚拟环境、容器镜像或发布产物
-3. 对近期安装过可疑版本的机器轮换 API Key 和相关凭证
-4. 升级到本热修复版本
+1. Check if `litellm` is installed in your environment
+2. Uninstall suspicious versions and rebuild virtual environments, images, or artifacts
+3. Rotate API keys and credentials on machines that installed suspicious versions
+4. Upgrade to this hotfix version
-可用命令:
+LiteLLM features will remain unavailable until a trusted upstream fix is released.
-## 兼容性说明
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.9...v0.2.10)
-这是一个以止损为目标的防御性热修复版本。LiteLLM 相关能力会暂时不可用,直到上游给出可信的修复版本和完整事故说明。
+## v0.2.9 (2026-03-19)
-## 参考链接
+This release focuses on stability and developer experience improvements. Key fixes address RocksDB lock contention by sharing a single adapter across account backends, restore previously lost plugin bug fixes, and improve vector store incremental updates. New features include a bot debug mode with `/remember` command support, summary-based file embedding in the semantic pipeline, and comprehensive PR-Agent review rules for CI. Documentation also received Docker Compose and Mac port forwarding guidance.
-- 上游 issue: <https://github.com/BerriAI/litellm/issues/24512>
-- PyPI 项目页: <https://pypi.org/project/litellm/>
-- GitHub Security: <https://github.com/BerriAI/litellm/security>
-- OpenViking 热修分支: <https://github.com/volcengine/OpenViking/tree/hotfix/v0.2.10-disable-litellm>
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.8...v0.2.9)
-## v0.2.9 (2026-03-19)
+## v0.2.8 (2026-03-19)
-## What's Changed
-* fix(resource): enforce agent-level watch task isolation
-* feat(embedder): use summary for file embedding in semantic pipeline
-* Fix/bot readme
-* Fix/increment update dir vector store
-* fix(plugin): restore bug fixes from #681 and #688 lost during #662 merge
-* docs: add docker compose instructions and mac port forwarding tip to …
-* Update docs
-* feat(ci): add comprehensive Qodo PR-Agent review rules
-* [feat](bot):Add mode config, add debug mode, add /remember cmd
-* fix(vectordb): share single adapter across account backends to prevent RocksDB lock contention
+### Highlights
-## v0.2.8 (2026-03-19)
+- OpenClaw plugin upgraded to 2.0 (context engine), OpenCode memory plugin added, multi-agent memory isolation via `agentId`.
+- Memory cold-storage archival with hotness scoring, chunked vectorization for long memories, `used()` tracking interface.
+- Rerank integration in hierarchical retrieval, RetrievalObserver for quality metrics.
+- Resource watch scheduling, reindex endpoint, legacy `.doc`/`.xls` parser support, path locking and crash recovery.
+- Request-level trace metrics, memory extract telemetry breakdown, OpenAI VLM streaming, `<think>` tag cleanup.
+- Cross-platform fixes (Windows zip, Rust CLI), AGFS Makefile refactor, CPU-variant vectordb engine, Python 3.14 wheel support.
-# OpenViking v0.2.8 发布公告
-
-## 本次更新亮点
-
-### 1. 插件生态继续升级,OpenClaw / OpenCode 集成更完整
-- `openclaw-plugin` 升级到 **2.0**,从 memory plugin 进一步演进为 **context engine**。
-- 新增并完善 **OpenCode memory plugin example**,补充 attribution 与后续插件更新。
-- 支持 **多智能体 memory isolation**,可基于 hook context 中的 `agentId` 做记忆隔离。
-- 修复插件自动 recall 可能导致 agent 长时间挂起的问题,并补强 legacy commit 兼容。
-- MCP 查询服务补充 `api_key` 支持与默认配置能力,降低接入成本。
-
-### 2. Session / Memory 链路增强,长期记忆能力更稳
-- 新增 **memory cold-storage archival**,通过 hotness scoring 管理长期记忆冷热分层。
-- 新增长记忆 **chunked vectorization**,改善超长内容的向量化处理能力。
-- 增加 `used()` 接口,用于上下文 / skill 使用追踪。
-- 修复 async commit 过程中的上下文透传、批次内重复写入、非标准 LLM 响应处理等问题。
-- Bot 场景下改用 `commit_async()`,减少阻塞风险,提升长会话稳定性。
-
-### 3. 检索与 Embedding 能力持续增强
-- 分层检索中正式集成 **rerank**,并修复无 rerank 场景下的检索可用性问题。
-- 新增 **RetrievalObserver**,可用于检索质量指标观测,并已接入 `ov observer`。
-- Embedding 侧新增:
-- 修复 CJK token 估算偏低等问题,提升多语言场景下的稳定性。
-
-### 4. 资源、存储与解析链路更完善
-- 新增 **resource watch scheduling** 与状态跟踪能力,资源同步流程更可控。
-- 新增 **reindex endpoint**,支持内容手动修改后的重新 embedding。
-- 解析能力新增对 **legacy `.doc` / `.xls`** 格式的支持。
-- 修复从 URL 导入文件时文件名与扩展名丢失的问题。
-- 存储层新增 **path locking** 与选择性 crash recovery,进一步提升写入安全性。
-- 修复 tenant API 的隐式 root fallback、文件系统接口路径处理、工作目录 `~` 展开等问题。
-
-### 5. VLM、Trace 与可观测性能力加强
-- 新增 **request-level trace metrics** 与对应 API 支持。
-- 新增 **memory extract telemetry breakdown**,帮助更细粒度地分析记忆提取过程。
-- OpenAI VLM 支持 **streaming response handling**。
-- 补充 `max_tokens` 参数以避免 vLLM 拒绝请求,并支持 OpenAI-compatible VLM 的自定义 HTTP headers。
-- 自动清理模型输出中的 `<think>` 标签,减少推理内容污染存储结果。
-
-### 6. 工程兼容性与交付体验继续改进
-- 修复 Windows zip 路径、代码仓库索引、Rust CLI 版本等跨平台问题。
-- `agfs` Makefile 完成跨平台兼容性重构。
-- Vectordb engine 支持按 **CPU variant** 拆分。
-- Docker 构建链路持续修复,补齐 `build_support/` 拷贝逻辑。
-- Release workflow 支持发布 **Python 3.14 wheels**。
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.6...v0.2.8)
## v0.2.6 (2026-03-11)
-# OpenViking v0.2.6 发布公告
+### Highlights
-## 重点更新
+- CLI UX: `ov chat` with `rustyline` line editing, Markdown rendering, chat history.
+- Async capabilities: session commit with `wait` parameter, configurable worker count.
+- New OpenViking Console web UI for debugging and API exploration.
+- Bot enhancements: eval support, `add-resource` tool, Feishu progress notifications.
+- OpenClaw memory plugin major upgrade: npm install, consolidated installer, stability fixes.
+- Platform: Linux ARM support, Windows UTF-8 BOM fix, CI runner OS pinning.
-### 1. CLI 与对话体验进一步升级
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.5...v0.2.6)
-- `ov chat` 现在基于 `rustyline` 提供更完整的行编辑体验,终端交互更自然,不再出现常见的方向键控制字符问题。
-- 新增 Markdown 渲染能力,终端中的回答展示更清晰,代码块、列表等内容可读性更好。
-- 支持聊天历史记录,同时提供关闭格式化和关闭历史记录的选项,便于在不同终端环境中按需使用。
+## v0.2.5 (2026-03-06)
-### 2. 服务端异步能力增强,长任务不再轻易阻塞
+### Highlights
-- Session commit 新增异步提交能力,并支持通过 `wait` 参数控制是否同步等待结果。
-- 当选择后台执行时,OpenViking 现在会返回可追踪的任务信息,调用方可以通过任务接口查询状态、结果或错误。
-- 服务端新增可配置 worker count,进一步缓解单 worker 场景下长任务阻塞请求的问题。
+- **PDF & Parsing**: Font-based heading detection and bookmark extraction as structured markdown headings; `add_resource` now supports index control with refactored embedding logic and correctly handles ZIP-based container formats.
+- **Session & Memory**: `add_message()` adds `parts` parameter support; semantic indexing triggered for parent directories after memory extraction.
+- **URI Handling**: Short-format `VikingURI` support, `git@` SSH URL format in the CLI, and GitHub `tree/<ref>` URL for code repository import.
+- **Bot & Integrations**: VikingBot refactored with new evaluation module, Feishu multi-user and channel enhancements, OpenAPI standardization; Telegram crash fix for Claude.
+- **Infrastructure**: `agfs` gains ripgrep-based grep acceleration and async grep support with optional binding client mode; automated PR review workflows using Doubao model with severity classification.
+- **Installation**: curl-based installation works correctly on Ubuntu/Debian without triggering system protection errors; Rust compile fixed for `uv pip install -e .`.
-### 3. 新增 Console,并持续增强 Bot 与资源操作能力
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.3...v0.2.5)
-- 本版本新增独立的 OpenViking Console,提供更直观的 Web 控制台入口,方便调试、调用和查看接口结果。
-- Bot 能力继续增强,新增 eval 能力,开放 `add-resource` 工具,并支持飞书进度通知等扩展能力。
-- 资源导入支持 `preserve_structure` 选项,扫描目录时可以保留原始目录层级,适合更复杂的知识组织方式。
-- 同时修复了 grep、glob、`add-resource` 等场景下的一些响应问题,提升日常使用稳定性。
+## v0.2.3 (2026-03-03)
-### 4. openclaw memory plugin 能力和稳定性大幅完善
+### Breaking Change
-- openclaw memory plugin 在这一版中获得了较大幅度增强,补充了更完整的插件能力与相关示例。
-- 插件安装链路进一步完善,现已支持通过 npm 包方式安装 setup-helper,部署和升级都更直接。
-- 修复了本地运行、端口管理、配置保留、缺失文件下载、配置覆盖等一系列影响安装和使用体验的问题。
-- Memory consolidation 与 skill tool memory 相关问题也得到修复,进一步提升记忆链路的稳定性。
-- Vikingbot 作为 OpenViking 可选依赖的集成方式也做了完善,降低了插件和 bot 协同使用时的接入门槛。
+Datasets and indexes generated by previous versions are incompatible with this release and cannot be reused. A full rebuild is required after upgrading to avoid retrieval anomalies, inconsistent filtering, or runtime errors. Stop the service, remove your workspace directory (`rm -rf ./your-openviking-workspace`), then restart with `openviking-server`.
-### 5. 安装、跨平台与工程质量继续补强
+This release delivers CLI optimizations including `glob -n` flag support and `cmd echo`, alongside README updates for both English and Chinese documentation.
-- 新增 Linux ARM 支持,进一步扩展了 OpenViking 的部署平台范围。
-- 修复了 Windows 下 UTF-8 BOM 配置文件的兼容问题,减少配置读取失败的情况。
-- 修复了 `install.sh`、`ov.conf.example`、setup 失败退出等问题,并补充了 openclaw memory plugin 的 npm 包安装路径,提升首次安装和配置过程的成功率。
-- CI 侧将 GitHub Actions runner 固定到明确 OS 版本,减少环境漂移带来的构建与发布不确定性。
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.2...v0.2.3)
-## 总结
+## v0.2.2 (2026-03-03)
-## What's Changed
-* feat: improve ov chat UX with rustyline and markdown rendering
-* fix: grep for binding-client
-* feat(server): add configurable worker count to prevent single-worker blocking
-* feat(sessions): add async commit support with wait parameter
+### Breaking Change
-## v0.2.5 (2026-03-06)
+Before upgrading, stop the VikingDB Server and clear your workspace directory. Indexes from prior versions are not forward-compatible with this release.
-## What's Changed
-* docs: use openviking-server to launch server
-* fix: Session.add_message() support parts parameter
-* feat: support GitHub tree/&lt;ref&gt; URL for code repository import
-* fix: improve ISO datetime parsing
-* feat(pdf): extract bookmarks as markdown headings for hierarchical parsing
-* feat: add index control to add_resource and refactor embedding logic
-* fix: support short-format URIs in VikingURI and VikingFS access control
-* fix: handle None data in skill_processor._parse_skill()
-* fix: fix add-resource
-* fix: treat only .zip as archive; avoid unzipping ZIP-based container …
-* fix(cli): support git@ SSH URL format in add-resource
-* feat(pdf): add font-based heading detection and refactor PDF/Markdown parsing
-* 支持通过curl方式安装部署openclaw+openviking插件
-* Feature/vikingbot_opt: OpenAPI interface standardization;Feishu multi-user experience; observability enhancements; configuration system modernization.
-* docs: pip upgrade suggestion
-* feat: define a system path for future deployment
-* chore: downgrade golang version limit, update vlm version to seed 2.0
-* [fix]: 修复通过curl命令安装场下下ubuntu/debian等系统触发系统保护无法安装的openviking的问题
-* fix(session): trigger semantic indexing for parent directory after memory extraction
-* feat(bot):Refactoring, New Evaluation Module, Feishu channel Opt & Feature Enhancements
-* chore: agfs-client默认切到binding-client模式
-* feat(agfs): add ripgrep-based grep acceleration and fix vimgrep line parser
-* [WIP] ci: add automated PR review workflow using Doubao model
-* fix(ci): use OPENAI_KEY instead of OPENAI.KEY in pr-review workflow
-* build(deps): bump actions/download-artifact from 7 to 8 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 6 to 7 by @dependabot[bot] in
-* build(deps): bump docker/login-action from 3 to 4 by @dependabot[bot] in
-* build(deps): bump docker/setup-qemu-action from 3 to 4 by @dependabot[bot] in
-* build(deps): bump docker/setup-buildx-action from 3 to 4 by @dependabot[bot] in
-* fix(packaging): sdist 排除运行时二进制
-* ci: enhance PR review with severity classification and checklist
-* enable injection for Collection and CollectionAdaptor
-* chore: 编译子命令失败报错, golang版本最低要求1.22+
-* feat(viking_fs) support async grep
-* OpenViking Plugin Exception Handling & Fixing
-* feat(agfs): make binding client optional, add server bootstrap, tune logging and CI
-* fix: rust compile in uv pip install -e .
-* [wip]Enhance OpenViking Status Checks in the OpenClaw Plugin
-* FIX: fixes multiple issues in the OpenViking chat functionality and unifies session ID generation logic between Python and Rust CLI implementations.
-* chore: 增加Makefile,方便build
-* fix(bot): fix Telegram channel init crash and empty content for Claude
-* fix: suport uv pip install openviking[bot]
-* fix: agfs-client默认使用http-client
-* fix(agfs): fix agfs binding-client import error
+This release adds C# AST extractor support for code parsing, fixes multi-tenant filtering, normalizes OpenViking memory target paths, and improves git repository detection with `git@` SSH URL support. The `agfs` dependency libraries and binaries are now pre-compiled, eliminating the need for a build step at install time. Documentation adds Qwen model usage instructions.
-## v0.2.3 (2026-03-03)
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.2.1...v0.2.2)
-## Breaking Change
-After upgrading, datasets/indexes generated by historical versions are not compatible with the new version and cannot be reused directly. Please rebuild the datasets after upgrading (a full rebuild is recommended) to avoid retrieval anomalies, inconsistent filtering results, or runtime errors.
-Stop the service -> rm -rf ./your-openviking-workspace -> restart the service with the openviking-server command.
+## v0.2.1 (2026-02-28)
-## What's Changed
-* Feat: CLI optimization
-* Update README.md
-* Update README_CN.md
-* feat: support glob -n and cmd echo
-* fix: fix release
+### Highlights
-## v0.2.2 (2026-03-03)
+- **Multi-tenancy**: Foundational multi-tenancy support at the API layer for isolated multi-user/team usage.
+- **Cloud-Native**: Cloud-native VikingDB support, improved cloud deployment docs and Docker CI.
+- **OpenClaw/OpenCode**: Official `openclaw-openviking-plugin` installation, `opencode` plugin introduction.
+- **Storage**: Vector database interface refactored, AGFS binding client, AST code skeleton extraction, private GitLab domain support.
+- **CLI**: `ov` command wrapper, `add-resource` enhancements, `ovcli.conf` timeout support, `--version` flag.
-## Breaking Change
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.18...v0.2.1)
-Warning: This Release includes Breaking Chage! Before upgrading, you should stop VikingDB Server and clear workspace dir first.
+## [email protected] (2026-02-27)
-## What's Changed
-* fix ci
-* 在readme补充千问使用方法
-* feat(parse): Add C# AST extractor support
-* chore: bump CLI version to 0.2.1
-* fix: normalize OpenViking memory target paths
-* fix: fix filter when multi tenants
-* docs/update_wechat
-* Fix retriever
-* fix(git): support git@ SSH URLs with regression-safe repo detection
-* fix(agfs): 预编译agfs所依赖的lib/bin,无需安装时构建
+Updated CLI binary release with cross-platform support for macOS and Linux, aligned with the v0.1.18 feature set including the Rust-based implementation and expanded file parser capabilities.
-## v0.2.1 (2026-02-28)
+[Full Changelog](https://github.com/volcengine/OpenViking/releases/tag/cli%400.2.0)
-## 1. Core Capability Upgrades: Multi-tenancy, Cloud-Native & OpenClaw/OpenCode Adaptation
-- **Multi-tenancy**: Implemented foundational multi-tenancy support at the API layer (#260, #283), laying the groundwork for isolated usage across multiple users/teams.
-- **Cloud-Native Support**: Added support for cloud-native VikingDB (#279), improved cloud deployment documentation and Docker CI workflows (#320), expanding from local deployment to cloud-edge integration.
-- **OpenClaw Integration**: Added official installation for the `openclaw-openviking-plugin` (#307), strengthening integration with OpenClaw.
-- **OpenCode Support**: Introduced the `opencode` plugin and updated documentation (#351), extending capabilities for code-related scenarios.
-
-## 2. Deep Optimization of the Database Storage Foundation
-- **Architecture Refactor**: Refactored the vector database interface (#327) and removed the `with_vector` parameter from query APIs (#338) to simplify the interface design.
-- **Performance Optimizations**:
-- **BugFixes**: Resolved missing Volcengine URI prefixes for VikingDB, invalidated `is_leaf` filters (#294), and fixed vector storage lock contention during fast restarts (#343).
-- **AGFS Enhancements**: Added AGFS binding client (#304), fixed AGFS SDK installation/import issues (#337, #355), and improved filesystem integration.
-- **Code Scenario Improvements**: Added AST-based code skeleton extraction mode (#334), supported private GitLab domains for code repositories (#285), and optimized GitHub ZIP download (#267).
-
-## 3. Improved CLI Toolchain (Greatly Enhanced Usability)
-- Added `ov` command wrapper (#325) and fixed bugs in the CLI wrapper, repo URI handling, and `find` command (#336, #339, #347).
-- Enhanced `add-resource` functionality with unit tests (#323) and added ZIP upload support for skills via the `add_skill` API (#312).
-- Configuration Extensions: Added timeout support in `ovcli.conf` (#308) and fixed agent_id issues in the Rust CLI (#308).
-- Version Support: Added the `--version` flag to `openviking-server` (#358) for easy version validation.
-
-## What's Changed
-* docs : update wechat
-* feat: 多租户 Phase 1 - API 层多租户能力
-* 增加openviking/eval模块,用于评估测试
-* feat(vectordb): integrate KRL for ARM Kunpeng vector search optimization
-* feat: concurrent embedding, GitHub ZIP download, read offset/limit
-* fix: claude code memory-plugin example:add_message改为写入TextPart列表,避免session解析异常
-* Feat/add parts support to http api
-* tests(parsers): add unit tests for office extensions within add_resou…
-* feat: break change, remove is_leaf scalar and use level instead
-* fix(api): complete parts support in SDK layers and simplify error handling
-* feat: support cloud vikingDB
-* Fix image_summary bug
-* feat(config): expose embedding.max_concurrent and vlm.max_concurrent …
-* Multi tenant
-* feat: allow private gitlab domain for code repo
-* fix(storage): idempotent rm/mv operations with vector index sync
-* refactor(filter): replace prefix operator with must
-* fix: VikingDB volcengine URI prefix loss and stale is_leaf filter
-* fix(vectordb): default x86 build to AVX2 and disable AVX512
-* 对齐 docs 和代码里 storage config 的 workspace 属性
-* fix: correct storage.update() call signature in _update_active_counts()
-* feat(memory): add hotness scoring for cold/hot memory lifecycle
-* feat(client): ovcli.conf 支持 timeout 配置 + 修复 Rust CLI agent_id
-* fix(server): 未配置 root_api_key 时仅允许 localhost 绑定
-* fix: resolve Gemini 404, directory collision, and Unicode decoding er…
+## v0.1.18 (2026-02-23)
-## [email protected] (2026-02-27)
+This release brings major new capabilities to OpenViking. A high-performance Rust CLI is introduced alongside a terminal UI for filesystem navigation. File parsing is significantly expanded with support for Word, PowerPoint, Excel, EPub, and ZIP formats. Multi-provider support is added for embedding and VLM backends. Memory handling is redesigned with conflict-aware deduplication and a new extraction flow.
-# OpenViking CLI v0.2.0
+### Highlights
-### Quick Install (macOS/Linux)
+- **Rust CLI**: New blazing-fast CLI implementation.
+- **File Parsers**: Word, PowerPoint, Excel, EPub, ZIP support via markitdown-inspired parsers.
+- **TUI**: Basic terminal UI for filesystem navigation (`ov tui`).
+- **Multi-provider**: Support for multiple embedding and VLM providers.
+- **Memory**: Redesigned extraction and deduplication flow with conflict awareness.
+- **Skills**: New memory, resource, and search skills; improved skill search ranking.
+- **Directory Parsing**: Added directory-level parsing support.
-### Manual Installation
-Download the appropriate binary for your platform below, extract it, and add it to your PATH.
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.17...v0.1.18)
-The CLI command is simply `ov`:
+## [email protected] (2026-02-14)
-### Checksums
-SHA256 checksums are provided for each binary for verification.
+Initial CLI binary release with cross-platform support for macOS and Linux, providing a standalone distributable for OpenViking server management and resource operations.
-## v0.1.18 (2026-02-23)
+[Full Changelog](https://github.com/volcengine/OpenViking/releases/tag/cli%400.1.0)
-## What's Changed
-* feat: add Rust CLI implementation [very fast]
-* feat: make -o and --json global param
-* feat: provide a test_ov.sh scripts as reference
-* fix: short markdown parse filename
-* Update 03-quickstart-server.md-en version
-* feat: add markitdown-inspired file parsers (Word, PowerPoint, Excel, EPub, ZIP)
-* feat: rename for consistency
-* 上传最新的微信群二维码
-* Update 03-quickstart-server.md
-* Update 01-about-us.md-微信群二维码地址更新
-* fix: build target name
-* fix: fix handle github url right
-* Update README_CN.md-新增云端部署跳转链接与说明
-* Update README.md-新增英文版readme云端部署内容
-* Update 01-about-us.md-英文版微信群二维码更新
-* fix: fix the difference of python CLI and rust CLI
-* Feat/support multi providers , OpenViking支持多providers
-* Fix: 修复 rust CLI 命令中与 python CLI ls/tree 命令不一致的部分;add node limit arg
-* feat: add_memory cli with ov-memory SKILL
-* feat: add directory parsing support to OpenViking
-* fix: auto-rename on duplicate filename conflict
-* fix: guard against None candidate in search_by_id
-* fix bugs 修复provider为openai,但api_key并不是https://api.openai.com/v1而引起的大模型调用失败情况
-* Update open_viking_config.py
-* fix(parser): hash & shorten filenames that exceed filesystem limit
-* test: add comprehensive edge case tests
-* fix: invalid Go version format in agfs-server go.mod
-* fix: add input validation to search_by_id method
-* fix: improve binary content detection and null byte handling
-* suggestion: align grep and glob options
-* suggestion: fix empty in --simple ls response
-* feat: support basic ov tui for fs navigator, boost version 0.2.0
-* build(deps): bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* build(deps): bump actions/checkout from 4 to 6 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 4 to 6 by @dependabot[bot] in
-* refactor(memory): redesign extraction/dedup flow and add conflict-a…
-* fix: target directories retrieve
-* fix: skill search ranking - use overview for embedding and fix visited set filtering
-* fix: use frontmatter description for skill vectorization instead of overview
-* build(deps): bump actions/cache from 4 to 5 by @dependabot[bot] in
-* feat: update media parsers
-* fix: update memory
-* fix: resolve -o option conflict between global output and filesystem output-format
-* feat: add memory, resource and search skills
+## v0.1.17 (2026-02-14)
-## [email protected] (2026-02-14)
+A stability-focused patch release. Reverts dynamic project name configuration in VectorDB due to instability, fixes CI workspace cleanup, and resolves a tree URI output error with added validation of `ov.conf` on startup.
-# OpenViking CLI v0.1.0
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.16...v0.1.17)
-### Quick Install (macOS/Linux)
+## v0.1.16 (2026-02-13)
-### Manual Installation
-Download the appropriate binary for your platform below, extract it, and add it to your PATH.
+A focused bug-fix and improvement release. Fixes VectorDB connectivity issues and a server conflict between uvloop and nest_asyncio. Temporary URIs are now human-readable, resource add timeouts are enlarged, and dynamic project name configuration is introduced for VectorDB and Volcengine backends.
-The CLI command is simply `ov`:
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.15...v0.1.16)
-### Checksums
-SHA256 checksums are provided for each binary for verification.
+## v0.1.15 (2026-02-13)
-## v0.1.17 (2026-02-14)
+This release focuses on architectural refactoring and reliability improvements. The HTTP client is split into distinct embedded and HTTP modes for cleaner separation of concerns. The CLI launch speed is improved through directory restructuring. VectorDB timestamp and collection creation bugs are resolved.
-## What's Changed
-* Revert "feat: support dynamic project_name config in VectorDB / volcengine"
-* Fix/ci clean workspace
-* fix: tree uri output error, and validate ov.conf before start
+### Highlights
-## v0.1.16 (2026-02-13)
+- **Refactor**: HTTP client split into embedded and HTTP modes; QueueManager decoupled from VikingDBManager.
+- **CLI**: Faster launch speed; improved `ls` and `tree` output.
+- **VectorDB**: Fixed timestamp format and collection creation issues.
+- **Parser**: Support for repository branch and commit refs.
+- **OpenClaw**: Initial memory output language pipeline adaptation.
-## What's Changed
-* fix: fix vectordb
-* feat: make temp uri readable, and enlarge timeout of add-resource
-* feat: support dynamic project_name config in VectorDB / volcengine
-* fix: server uvloop conflicts with nest_asyncio
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.14...v0.1.15)
-## v0.1.15 (2026-02-13)
+## v0.1.14 (2026-02-12)
-## What's Changed
-
-Now you can try Server/CLI mode!
-
-* refactor(client): 拆分 HTTP 客户端,分离嵌入模式与 HTTP 模式
-* Transaction store
-* fix CI: correct patch targets in test_quick_start_lite.py
-* fix/lifecycle
-* Fix/lifecycle
-* refactor: decouple QueueManager from VikingDBManager
-* refactor: to accelerate cli launch speed, refactor openviking_cli dir
-* efactor: to accelerate cli launch speed, refactor openviking_cli dir
-* fix(vectordb): resolve timestamp format and collection creation issues
-* doc: add multi-tenant-design
-* fix: fix vectordb sparse
-* [WIP]adapt for openclaw: add memory output language pipeline
-* CLI commands (ls, tree) output optimize
-* fix: replace bare except with Exception in llm utils
-* feat(parser): support repo branch and commit refs
-* fix: temp dir check failed
+A major infrastructure release. An HTTP server and Python HTTP client are introduced, enabling remote access to OpenViking services. The OpenClaw skill adds MCP integration support. Directory pre-scan validation, DAG-triggered embedding, and parallel resource addition improve performance and reliability.
-## v0.1.14 (2026-02-12)
+### Highlights
+
+- **HTTP Server**: New server mode with Python HTTP client for remote access.
+- **OpenClaw Skill**: MCP integration for OpenViking.
+- **CLI**: Full Bash CLI framework with comprehensive command implementation.
+- **Embedding**: DAG-triggered embedding and parallel add support.
+- **Directory Scan**: Pre-scan validation module added.
+- **Config**: Default configuration directory set to `~/.openviking`.
-## What's Changed
-* build(deps): bump protobuf from 6.33.2 to 6.33.5 by @dependabot[bot] in
-* refactor: cpp bytes rows
-* fix: agfs port
-* refactor: refactor agfs s3 backend config
-* fix(depends): 修复pip依赖的安全漏洞
-* feat: add HTTP Server and Python HTTP Client (T2 & T4)
-* Add OpenClaw skill for OpenViking MCP integration
-* docs: update docs and github workflows python version, python>=3.10
-* Doc: suggest use ~/.openviking/ov.conf as a default configure path
-* Parallel add
-* feat(directory-scan): add directory pre-scan validation module with f…
-* fix(docs): fix relative paths in README files to match actual docs st…
-* feat: use a default dir ~/.openviking to store configuration, fix win…
-* feat: dag trigger embedding
-* fix: fix windows
-* fix: release py3.13
-* fix: remove await asyncio and call agfs directly
-* fix(depends): 修复测试引入的依赖
-* add_message改为写入TextPart列表,避免session消息解析异常; extract_session增加json转换方法
-* feat: 新增 Bash CLI 基础框架与完整命令实现 (T3 + T5)
-* fix: test case
-* fix: fix release
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.12...v0.1.14)
## v0.1.12 (2026-02-09)
-## What's Changed
-* feat: add search_with_sparse_logit_alpha
-* refactor: Refactor S3 configuration structure and fix Python 3.9 compatibility issues
-* fix: fix ci
-* refactor: unify async execution utilities into run_async
-* docs: update community link
-* build(deps): bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* build(deps): bump actions/github-script from 7 to 8 by @dependabot[bot] in
-* build(deps): bump actions/checkout from 4 to 6 by @dependabot[bot] in
-* build(deps): bump actions/setup-go from 5 to 6 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 4 to 6 by @dependabot[bot] in
-* feat: in chatmem example, add /time and /add_resource command
-* Feature/new demo
-* Feat:支持原生部署的vikingdb
-* WIP: fix: run memex locally for #78
-* feat(parse): extract shared upload utilities
-* docs: fiix related document links in /openviking/parse/parsers/README.md
-* fix(parse): prevent Zip Slip path traversal in _extract_zip (CWE-22)
-* Path filter
-* feat: use tabulate for observer
-* fix(parser): fix temporary file leak in HTMLParser download link hand…
-* perf: reuse query embeddings in hierarchical retriever
-* fix(agfs): close socket on error path in _check_port_available
-* fix(storage): make VikingFS.mkdir() actually create the target directory
-* fix: fix sparse
-* feat: support query in mcp, tested with kimi
-* fix: ignore TestVikingDBProject
+This release improves search quality, storage reliability, and code maintainability. Sparse logit alpha search is added for enhanced retrieval. Query embeddings are reused in the hierarchical retriever for better performance. Native VikingDB deployment is now supported. A critical Zip Slip path traversal vulnerability (CWE-22) is patched.
+
+### Highlights
+
+- **Search**: Sparse logit alpha support and optimized query embedding reuse.
+- **VikingDB**: Native deployment support.
+- **Security**: Zip Slip path traversal fix (CWE-22).
+- **Refactor**: Unified async execution utilities; restructured S3 configuration.
+- **MCP**: Query support added and validated with Kimi.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.11...v0.1.12)
## v0.1.11 (2026-02-05)
-## What's Changed
-* support small github code repos
+Adds support for ingesting small GitHub code repositories, enabling OpenViking to index and search public codebases directly.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.10...v0.1.11)
## v0.1.10 (2026-02-05)
-## What's Changed
-* Fix compile
-* fix: fix windows release
+Patch release fixing a compilation error and resolving a Windows binary release packaging issue.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/compare/v0.1.9...v0.1.10)
## v0.1.9 (2026-02-05)
-## What's Changed
-* Bump github/codeql-action from 3 to 4 by @dependabot[bot] in
-* Bump actions/setup-go from 5 to 6 by @dependabot[bot] in
-* Bump actions/setup-python from 5 to 6 by @dependabot[bot] in
-* Bump astral-sh/setup-uv from 4 to 7 by @dependabot[bot] in
-* Bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* fix session test
-* feat: add GitHub issue and PR templates
-* fix: fix_build
-* docs: update readme
-* fix: Downgraded a log message from warning to info when the Rerank cl…
-* docs: update_readme
-* Remove agfs
-* docs: faq
-* fix: fix agfs-server for windows
-* Upgrade go
-* feat: add more visual example `query`
-* fix: support intel mac install
-* feat: rename backend to provider for embedding and vlm
-* Lint code
-* refactor: optimized pyproject.toml with optional groups
-* feat: linux compile
-* Change 'provider' to 'backend' in configuration
-* update version
-* Revert "Change 'provider' to 'backend' in configuration"
-* chore: use standard logging pkg
-* feat: chat & chat w/ mem examples
-* feat:常规环境下,增加python3.13适配
-* docs: add server cli design
-* refactor: refactor ci action
-* refactor: extract Service layer from async_client
-* fix: fix ci action
-* fix: simplify memory dedup decisions and fix retrieval recursion bug
-* Fix ci action
-* fix: 修复s3fs适配
-* refactor: extract ObserverService from DebugService for cleaner status access API
-* fix: fix release action
+The initial public release of OpenViking. This release establishes the core project structure with cross-platform support including Linux and Intel Mac. It introduces the service layer architecture, separating embedding and VLM backends into configurable providers. Memory deduplication is improved and retrieval recursion bugs are fixed. Python 3.13 compatibility, S3FS support, and usage examples for chat and memory workflows are included.
+
+### Highlights
+
+- **Initial Release**: Core OpenViking server, client, and CLI foundation.
+- **Providers**: Configurable embedding and VLM backends with provider abstraction.
+- **Architecture**: Service layer extracted from async client; ObserverService separated from DebugService.
+- **Platform**: Linux compile support, Intel Mac compatibility, Python 3.13 support.
+- **Memory**: Simplified deduplication logic and fixed retrieval recursion bug.
+- **Examples**: Chat and chat-with-memory usage examples added.
+
+[Full Changelog](https://github.com/volcengine/OpenViking/releases/tag/v0.1.9)
diff --git a/docs/en/about/03-roadmap.md b/docs/en/about/03-roadmap.md
index 7719966b2..0f913737d 100644
--- a/docs/en/about/03-roadmap.md
+++ b/docs/en/about/03-roadmap.md
@@ -9,66 +9,102 @@ This document outlines the development roadmap for OpenViking.
- Viking URI addressing system
- Dual-layer storage (AGFS + Vector Index)
- Async/Sync client support
+- QueueFS with SQLite backend
### Resource Management
- Text resource management (Markdown, HTML, PDF)
- Automatic L0/L1 generation
- Semantic search with vector indexing
- Resource relations and linking
+- Content write API
+- Agent namespace management
+
+### Multi-modal Parsing
+- Image OCR and parsing
+- Audio transcription (Whisper ASR)
+- Video parsing
+- PDF with bookmark extraction
+- Word, PowerPoint, Excel, EPub, ZIP parsers
+- Code file parsing
+- Feishu/Lark document parser
### Retrieval
- Basic semantic search (`find`)
- Context-aware search with intent analysis (`search`)
- Session-based query expansion
-- Reranking pipeline
+- Reranking pipeline with multiple providers (OpenAI, LiteLLM, Cohere, Volcengine)
-### Session Management
+### Session & Memory
- Conversation state tracking
- Context and skill usage tracking
- Automatic memory extraction
- Memory deduplication with LLM
- Session archiving and compression
+- Working Memory V2 with cold-storage archival
### Skills
- Skill definition and storage
- MCP tool auto-conversion
- Skill search and retrieval
-### Configuration
-- Pluggable embedding providers
+### Multi-tenant & Security
+- Multi-tenant support with account isolation
+- File and document encryption
+- User-level privacy configs API
+- API Key authentication
+
+### Configuration & Providers
+- Pluggable embedding providers (OpenAI, Gemini, Volcengine, MiniMax, LiteLLM, Jina, Cohere, DashScope, Voyage, local)
- Pluggable LLM providers
+- Pluggable rerank providers
- YAML-based configuration
+- Setup wizard (`openviking-server init`)
### Server & Client Architecture
- HTTP Server (FastAPI)
+- Native MCP endpoint built into openviking-server
- Python HTTP Client
-- API Key authentication
- Client abstraction layer (LocalClient / HTTPClient)
+- Web Console
+
+### CLI
+- Rust CLI (`ov` command)
+- TUI filesystem navigator
+- Privacy, search, session, resource, and admin commands
+
+### Bot Integration
+- VikingBot framework
+- Feishu/Lark channel
+- Telegram channel
+
+### Ecosystem & Plugins
+- OpenClaw plugin (context engine for coding agents)
+- Claude Code memory plugin
+- Codex memory plugin
+
+### Observability
+- Prometheus metrics
+- OpenTelemetry tracing
+- HTTP observability middleware
+
+### Deployment
+- Docker image and Docker Compose
+- Helm Chart for Kubernetes
+- Cloud VikingDB support
---
## Future Plans
-### CLI
-- Complete command-line interface for all operations
-- Distributed storage backend
-
-### Multi-modal Support
-- Intelligent parsing and access for multi-modal resources (images, video, audio, etc.)
-- Directory storage structure for multi-modal resources
-
### Context Management
- Propagation updates when context is modified
- Version management and rollback for context (git-like)
-### Resource Node Access Control
-- Multi-Agent / Multi-User support
-- Role-based isolation design
-- Access control and permission design for resource directory nodes
+### Distributed Storage
+- Distributed storage backend
-### Ecosystem Integration
-- Popular Agent framework adapters
-- Plugin system for custom components
+### Ecosystem
+- Additional Agent framework adapters
We welcome suggestions and feedback in issues.
diff --git a/docs/zh/about/01-about-us.md b/docs/zh/about/01-about-us.md
index 4042207c9..3170557da 100644
--- a/docs/zh/about/01-about-us.md
+++ b/docs/zh/about/01-about-us.md
@@ -1,4 +1,4 @@
-# OpenViking 团队与组织
+# 团队与组织
## 项目概述
diff --git a/docs/zh/about/02-changelog.md b/docs/zh/about/02-changelog.md
index c3ef2b281..57da2df84 100644
--- a/docs/zh/about/02-changelog.md
+++ b/docs/zh/about/02-changelog.md
@@ -3,39 +3,58 @@
OpenViking 的所有重要变更都将记录在此文件中。
此更新日志从 [GitHub Releases](https://github.com/volcengine/OpenViking/releases) 自动生成。
+## v0.3.14 (2026-04-30)
+
+### 重点更新
+
+- **可观测性**:OTLP 导出支持自定义 `headers`,覆盖 traces、logs、metrics 三条链路,便于直连需要额外鉴权头或 gRPC metadata 的观测后端。
+- **上传**:本地目录扫描和上传现在遵循根目录及子目录中的 `.gitignore` 规则,减少构建产物和临时文件被误导入。
+- **检索**:`search` / `find` 支持一次传入多个 target URI,适合跨目录、跨仓库范围检索。
+- **多租户**:OpenClaw 插件明确 `agent_prefix` 仅作为前缀使用;OpenCode memory plugin 补上 tenant headers 透传。
+- **管理**:新增 agent namespace 发现能力,服务端 API、CLI 和文档同步支持列出指定 account 下已有的 agent namespace。
+
+### 升级说明
+
+- OTLP 后端接入可通过 `headers` 统一配置鉴权信息(gRPC 模式为 metadata,HTTP 模式为请求头)。
+- 本地目录上传默认遵循 `.gitignore` 规则,此前被导入的临时/生成文件升级后可能被自动过滤。
+- OpenClaw 插件 `agent_prefix` 仅表示前缀,文档中 `agentId` 已统一迁移为 `agent_prefix`。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.13...v0.3.14)
+
+## v0.3.13 (2026-04-29)
+
+### 重点更新
+
+- **内置 MCP 端点**:`openviking-server` 在同一进程、同一端口暴露 `/mcp`,复用 REST API 的 API-Key 鉴权,提供 `search`、`read`、`list`、`store`、`add_resource`、`grep`、`glob`、`forget`、`health` 9 个工具。
+- **用户级隐私配置**:新增 `/api/v1/privacy-configs` API 和 `openviking privacy` CLI,按 `category + target_key` 保存、轮换、回滚 skill 等敏感配置。
+- **可观测性升级**:统一 `server.observability` 配置,支持 Prometheus `/metrics` 和 OpenTelemetry metrics/traces/logs 导出。
+- **检索调优**:新增 `embedding.text_source`、`embedding.max_input_tokens`、`retrieval.hotness_alpha`、`retrieval.score_propagation_alpha` 等配置。
+- **API 语义收敛**:搜索空 query 提前拒绝;公开 `viking://` URI 校验更严格;错误统一进入标准 error envelope。
+- **Docker 体验**:持久化状态收敛到 `/app/.openviking`;缺少 `ov.conf` 时容器存活并返回 503 初始化指引。
+- **安全**:bot 图片工具禁止读取沙箱外文件;health check 无凭证时跳过身份解析;API key 字段哈希拆分为独立开关。
+
+### 升级说明
+
+- `encryption.api_key_hashing.enabled` 需要显式配置(默认 `false`)。如依赖旧的隐式哈希行为,需手动开启。
+- OpenClaw 插件仅保留远程模式,不再启动本地子进程;`agentId` → `agent_prefix`,`recallTokenBudget` → `recallMaxInjectedChars`。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.12...v0.3.13)
+
## v0.3.12 (2026-04-24)
-## What's Changed
-* feat(semantic): add output_language_override to pin summary/overview language
-* docs(bot): sync gateway config example with #1640 security defaults
-* fix(parser): validate feishu config limits
-* fix(code-hosting): recognize SSH repository hosts with userinfo
-* docs(design): sync §4.6 role_id rules with #1643 passthrough behavior
-* feat: 补充百炼记忆库 LoCoMo benchmark 评测脚本
-* docs(contributing): add maintainer routing map
-* Revert "feat: 补充百炼记忆库 LoCoMo benchmark 评测脚本"
-* [fix] Fix/oc2ov test stability
-* fix(server): map AGFS URI errors
-* feat(git): added azure devops support
-* feat: support trusted admin tenant management
-* fix(session): count tool parts in pending tokens
-* fix(user_id): support dot
-* Add VitePress docs site and Pages deployment
-* feat(parse): support larkoffice.com Feishu document URLs
-* feat(ragfs): add s3 key normalization encoding
-* fix: apikey security: API Key 管理重构与安全增强
-* add Copy Markdown button and llms.txt support
-* fix(api_keys): proxy get_user_role in NewAPIKeyManager (trusted-mode 500 regression)
-* Fix docs npm registry for release builds
-* fix: account name security
+### 重点更新
-## v0.3.10 (2026-04-23)
+- **新集成**:新增 Azure DevOps Git 托管支持和 larkoffice.com 飞书文档 URL 解析。
+- **安全**:API Key 管理重构与安全增强,修复 account name 暴露问题,解决 trusted-mode proxy role 查询 500 回退。
+- **文档**:上线 VitePress 文档站并部署到 GitHub Pages,新增 llms.txt 支持和 Copy Markdown 按钮。
+- **Bug 修复**:修正飞书 config 限制校验、SSH 仓库 host 的 userinfo 识别、AGFS URI 错误映射、pending tool parts token 计数。
+- **开发者体验**:新增 maintainer routing map 贡献文档,RAGFS 新增 S3 key normalization encoding。
-### 版本概览
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.10...v0.3.12)
-v0.3.10 重点增强了 VLM provider、OpenClaw 插件生态、VikingDB 数据面接入,以及文件写入、QueueFS、Bot/CLI 的稳定性。本次发布包含 46 个提交,覆盖新功能、兼容性修复、安全修复和测试补强。
+## v0.3.10 (2026-04-23)
-### 主要更新
+### 重点更新
- 新增 Codex、Kimi、GLM VLM provider,并支持 `vlm.timeout` 配置。
- 新增 VikingDB `volcengine.api_key` 数据面模式,可通过 API Key 访问已创建好的云上 VikingDB collection/index。
@@ -60,853 +79,376 @@ v0.3.10 重点增强了 VLM provider、OpenClaw 插件生态、VikingDB 数据
- 修复 Gemini + tools 场景下 LiteLLM `cache_control` 导致的 400 错误,并支持 OpenAI reasoning model family。
- 修复 S3FS 目录 mtime 稳定性、Rust native build 环境污染、SQLite 数据库扩展名解析等问题。
-### 文档、测试与安全
-
-- 补充 VLM provider、Codex OAuth、Kimi/GLM、`write(mode=create)`、`tools.mcp_servers`、`ov_tools_enable`、Feishu thread 和 VLM timeout 文档。
-- 新增资源构建、Context Engine、OpenClaw 插件、内容写入、VLM provider、setup wizard、server bootstrap 和安全相关测试。
-- 新增 `SECURITY.md` 并更新 README、多语言文档和社群二维码。
-- 修复多项 code scanning 和 runtime 安全告警。
-- 增强 Bot gateway、OpenAPI auth、werewolf demo、配置校验和本地命令执行相关安全测试。
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.9...v0.3.10)
## v0.3.9 (2026-04-18)
-## What's Changed
-* reorg: remove golang depends
-* Feat/mem opt
-* fix: openai like embedding models fix, no more matryoshka error
-* feat(bot): Add disable OpenViking config for channels.
-* fix(config): point missing-config help messages to openviking.ai docs
-* fix(embedder): initialize async client state in VolcengineSparseEmbedder
-* feat(examples): add Codex memory plugin example
-* feat(openclaw-plugin): add unified ov_import and ov_search
-* feat(bot): add MCP client support (port from HKUDS/nanobot v0.1.5)
-* feat(eval):Readme add qa
-* feat(cli): support for default file/dir ignore config in `ovcli.conf`
-* benchmark: add LoCoMo evaluation for Supermemory
-* fix(embedder): report configured provider in slow-call logs
-* fix(queue): preserve embedding message ids across serialization
-* test(security): add unit tests for network_guard and zip_safe modules
-* fix(semantic): preserve repository hierarchy in overviews
-* fix(tests): align pytest coverage docs with required setup
-* feat: rerank support extra headers
-* fix: reload legacy session rows
-* fix: protect global watch-task control files from non-root access
-* fix(agfs): enable agfs s3 plugin default
-* fix(claude-code-memory-plugin): improve Windows compatibility
-* fix(pdf): resolve bookmark page mapping
-* fix: update observer test to use /models endpoint instead of non-exis…
-* fix(openclaw-plugin): extend default Phase 2 commit wait timeout
-* pref(retrieve): Optimize the search performance of larger directories by skipping redundant target_directories scope
-* Add third_party directory to Dockerfile
-* Fix/openclaw addmsg
-* feat(bot):Heartbeat fix
-* feat: add `openviking-server init` interactive setup wizard for local Ollama model deployment
-* fix(volcengine): update default doubao embedding model
-* feat: add Memory V2 full suite test
-* update new wechat group qr code
-* feat(filesystem): support directory descriptions on mkdir
-* feat(memory): default to memory v2
-* fix: resolve OpenClaw session file lock conflicts in oc2ov tests
-* fix: isolate temp scope by user within an account
-* fix(docker): raise Rust toolchain for ragfs image builds
-* feat(metric): add metric system
-* openclaw refactor: assemble context partitioning (Instruction/Archive/Session/…
-* Fix/memory v2
-* Fix/memory v2
-* reorg: split parser layer to 2-layer: accessor and parser, so that we can reuse more code
-* fix: merge error
+### 重点更新
+
+- **Memory**:Memory V2 设为默认,包含完整测试套件、session 行迁移,修复并发场景下的文件锁冲突。
+- **OpenClaw**:上下文分区重构为 Instruction/Archive/Session 层,插件统一 `ov_import` 和 `ov_search`,延长 Phase 2 commit 等待超时。
+- **Bot & MCP**:从 HKUDS/nanobot v0.1.5 移植 MCP client 支持,新增单 channel 禁用 OpenViking 配置,修复心跳可靠性。
+- **检索与搜索**:通过跳过冗余 scope 检查优化大目录搜索性能,修复 sparse embedder 异步初始化,新增 rerank extra-headers 支持。
+- **部署与上手**:新增交互式 `openviking-server init` 向导支持本地 Ollama 部署,`ovcli.conf` 新增默认文件/目录忽略配置。
+- **基础设施**:新增度量系统,更新默认 Doubao embedding 模型,提升 RAGFS Docker 构建的 Rust toolchain,解析器拆分为 accessor 和 parser 两层。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.8...v0.3.9)
## v0.3.8 (2026-04-15)
### Memory V2 专题
-- 记忆格式:
-- 重构与优化:
-- 对用户的直接价值:
+Memory V2 现在作为默认记忆管线,采用全新格式、重构的抽取与去重流程,长期记忆质量显著提升。
### 重点更新
-- Memory V2 默认开启:
-- 本地部署与初始化体验:
-- 插件与 Agent 生态增强:
-- 配置与部署体验改进:
-- 性能与稳定性:
-- 其他补充:
+- Memory V2 默认开启,格式与抽取管线全面重构。
+- 本地部署与初始化体验增强(`openviking-server init`)。
+- 插件与 Agent 生态增强(Codex、OpenClaw、OpenCode 示例)。
+- 配置与部署体验改进(S3 批量删除开关、OpenRouter `extra_headers`)。
+- Memory、Session、存储层性能与稳定性改进。
### 升级提示
-- 如果你经常通过 CLI 导入目录资源,建议在 `ovcli.conf` 中配置 `upload.ignore_dirs`,减少无关目录上传。
-- 如果你需要保留旧行为,可在 `ov.conf` 中显式设置 `"memory": { "version": "v1" }` 回退到 legacy memory pipeline。
-- 如果你之前使用 `ov init` 或 `ov doctor`,请改用 `openviking-server init` 和 `openviking-server doctor`。
-- 如果你使用 OpenRouter 或其他 OpenAI 兼容 rerank/VLM 服务,可以通过 `extra_headers` 注入平台要求的 Header。
-- 如果你的对象存储是阿里云 OSS 或其他 S3 兼容实现,且批量删除存在兼容问题,可开启 `storage.agfs.s3.disable_batch_delete`。
-- 如果你在做 Agent 集成,建议查看 `examples/codex-memory-plugin` 与 `examples/openclaw-plugin` 中的新示例和工具能力。
+- 如果你经常通过 CLI 导入目录资源,建议在 `ovcli.conf` 中配置 `upload.ignore_dirs`。
+- 旧行为可通过 `"memory": { "version": "v1" }` 回退。
+- `ov init` / `ov doctor` 请改用 `openviking-server init` / `openviking-server doctor`。
+- OpenRouter 或其他 OpenAI 兼容 rerank/VLM 服务可通过 `extra_headers` 注入平台要求的 Header。
+- S3 兼容实现批量删除有兼容问题时,可开启 `storage.agfs.s3.disable_batch_delete`。
-## v0.3.5 (2026-04-10)
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.5...v0.3.8)
-## What's Changed
-* fix(memory): define config before v2 memory lock retry settings access
-* fix: 优化测试关键词匹配和移除 Release Approval Gate
-* fix: sanitize internal error details in bot proxy responses
-* feat: add scenario-based API tests
-* fix(security): remove leaked token from settings.py
-* Fix/add resource cover
-* Revert "Fix/add resource cover"
-* fix: litellm embedding dimension adapts
-* ci: optimize runner usage with conditional OS matrix and parallel limit
-* fix(bot):Response language, Multi user memory commit
-* ci: remove lite and full test workflows
-* docs: fix docker deployment
-* docs(openclaw-plugin): add health check tools guide
-* fix(queue): expose re-enqueue counts in queue status
-* feat(s3fs): add disable_batch_delete option for OSS compatibility
-* Fix/add resource cover
-* afterTurn: store messages with actual roles and skip heartbeat messages
-* fix: fall back to prefix filters for volcengine path scope
-* fix(session): auto-create missing sessions on first add
-* Fix/api test issues
-* fix: derive context_type from URI in index_resource
+## v0.3.5 (2026-04-10)
-## v0.3.4 (2026-04-09)
+### 重点更新
-# OpenViking v0.3.4
-
-## 版本亮点
-
-- **OpenClaw 插件与评测体验继续完善**:调整 `recallPreferAbstract` 与 `ingestReplyAssist` 的默认值以降低意外行为,[PR #1204](https://github.com/volcengine/OpenViking/pull/1204) [PR #1206](https://github.com/volcengine/OpenViking/pull/1206);新增 OpenClaw eval shell 脚本并修复评测导入问题,[PR #1287](https://github.com/volcengine/OpenViking/pull/1287) [PR #1305](https://github.com/volcengine/OpenViking/pull/1305);同时补齐 autoRecall 搜索范围与查询清洗、截断能力,[PR #1225](https://github.com/volcengine/OpenViking/pull/1225) [PR #1297](https://github.com/volcengine/OpenViking/pull/1297)。
-- **Memory、会话写入与运行时稳定性明显增强**:写接口引入 request-scoped wait 机制,[PR #1212](https://github.com/volcengine/OpenViking/pull/1212);补强 PID lock 回收、召回阈值绕过、孤儿 compressor 引用、async contention 和 memory 语义批处理等问题,[PR #1211](https://github.com/volcengine/OpenViking/pull/1211) [PR #1301](https://github.com/volcengine/OpenViking/pull/1301) [PR #1304](https://github.com/volcengine/OpenViking/pull/1304);并继续优化 memory v2 compressor 锁重试控制,[PR #1275](https://github.com/volcengine/OpenViking/pull/1275)。
-- **安全与网络边界进一步收紧**:HTTP 资源导入补齐私网 SSRF 防护,[PR #1133](https://github.com/volcengine/OpenViking/pull/1133);trusted mode 在无 API key 时被限制为仅允许 localhost,[PR #1279](https://github.com/volcengine/OpenViking/pull/1279);embedding circuit breaker 与日志抑制也变得可配置,[PR #1277](https://github.com/volcengine/OpenViking/pull/1277)。
-- **生态与集成能力继续扩展**:新增 Volcengine Vector DB STS Token 支持,[PR #1268](https://github.com/volcengine/OpenViking/pull/1268);新增 MiniMax-M2.7 与 MiniMax-M2.7-highspeed provider 支持,[PR #1284](https://github.com/volcengine/OpenViking/pull/1284);AST 侧补充 Lua parser,[PR #1286](https://github.com/volcengine/OpenViking/pull/1286);Bot 新增 channel mention 能力,[PR #1272](https://github.com/volcengine/OpenViking/pull/1272)。
-- **发布、Docker 与 CI 链路更稳健**:发布时自动更新 `main` 并增加 Docker Hub push,[PR #1229](https://github.com/volcengine/OpenViking/pull/1229);修复 Docker maturin 路径并将 Gemini optional dependency 纳入镜像,[PR #1295](https://github.com/volcengine/OpenViking/pull/1295) [PR #1254](https://github.com/volcengine/OpenViking/pull/1254);CI 侧优化 API matrix、补充 timeout/SMTP 通知并修复 reusable workflow / action 问题,[PR #1281](https://github.com/volcengine/OpenViking/pull/1281) [PR #1293](https://github.com/volcengine/OpenViking/pull/1293) [PR #1300](https://github.com/volcengine/OpenViking/pull/1300) [PR #1302](https://github.com/volcengine/OpenViking/pull/1302) [PR #1307](https://github.com/volcengine/OpenViking/pull/1307)。
-
-## 升级说明
-
-- OpenClaw 插件默认配置发生调整:`recallPreferAbstract` 与 `ingestReplyAssist` 现在默认均为 `false`。如果你之前依赖默认开启行为,升级后需要显式配置,见 [PR #1204](https://github.com/volcengine/OpenViking/pull/1204) 和 [PR #1206](https://github.com/volcengine/OpenViking/pull/1206)。
-- HTTP 资源导入现在默认更严格地防护私网 SSRF;如果你有合法的内网资源采集场景,升级时建议复核现有接入方式与白名单策略,见 [PR #1133](https://github.com/volcengine/OpenViking/pull/1133)。
-- trusted mode 在未提供 API key 时已限制为仅允许 localhost 访问;如果你此前通过非本地地址使用 trusted mode,需要同步调整部署与鉴权配置,见 [PR #1279](https://github.com/volcengine/OpenViking/pull/1279)。
-- 写接口现在引入 request-scoped wait 机制,相关调用在并发写入下的等待与返回时机会更一致;如果你有依赖旧时序的外部编排逻辑,建议升级后复核行为,见 [PR #1212](https://github.com/volcengine/OpenViking/pull/1212)。
-- server 已支持 `host=none` 以使用双栈网络;如果你在 IPv4/IPv6 混合环境中部署,可考虑调整监听配置,见 [PR #1273](https://github.com/volcengine/OpenViking/pull/1273)。
-- Docker 镜像已纳入 Gemini optional dependency,并修复了 maturin 路径问题;如果你维护自定义镜像或发布流水线,建议同步检查构建脚本,见 [PR #1254](https://github.com/volcengine/OpenViking/pull/1254) 和 [PR #1295](https://github.com/volcengine/OpenViking/pull/1295)。
-
-### OpenClaw 插件、评测与集成
-
-- 默认将 openclaw-plugin 的 `recallPreferAbstract` 设为 `false`,[PR #1204](https://github.com/volcengine/OpenViking/pull/1204) by @wlff123
-- 修复 eval 中的 async import 问题,[PR #1203](https://github.com/volcengine/OpenViking/pull/1203) by @yeshion23333
-- 默认将 openclaw-plugin 的 `ingestReplyAssist` 设为 `false`,[PR #1206](https://github.com/volcengine/OpenViking/pull/1206) by @wlff123
-- 向 OpenAIVLM client 增加 timeout 参数,[PR #1208](https://github.com/volcengine/OpenViking/pull/1208) by @highland0971
-- 防止 redo recovery 中阻塞式 VLM 调用导致启动 hang 住,[PR #1226](https://github.com/volcengine/OpenViking/pull/1226) by @mvanhorn
-- 为插件 autoRecall 增加 `skills` 搜索范围,[PR #1225](https://github.com/volcengine/OpenViking/pull/1225) by @mvanhorn
-- 新增 bot channel mention 能力,[PR #1272](https://github.com/volcengine/OpenViking/pull/1272) by @yeshion23333
-- 新增 OpenClaw eval shell 脚本,[PR #1287](https://github.com/volcengine/OpenViking/pull/1287) by @yeshion23333
-- 新增 `add_message` 的 create time,[PR #1288](https://github.com/volcengine/OpenViking/pull/1288) by @wlff123
-- 对 OpenClaw recall query 做清洗并限制长度,[PR #1297](https://github.com/volcengine/OpenViking/pull/1297) by @qin-ctx
-- 修复 OpenClaw eval 导入到 OpenViking 时默认 user 的问题,[PR #1305](https://github.com/volcengine/OpenViking/pull/1305) by @yeshion23333
-
-### Memory、会话、存储与运行时
-
-- 修复明文文件长度小于 4 字节时 decrypt 抛出 `Ciphertext too short` 的问题,[PR #1163](https://github.com/volcengine/OpenViking/pull/1163) by @yc111233
-- 为短明文加密补充单元测试,[PR #1217](https://github.com/volcengine/OpenViking/pull/1217) by @baojun-zhang
-- 修复 PID lock 回收、召回阈值绕过与 orphaned compressor refs 等问题,[PR #1211](https://github.com/volcengine/OpenViking/pull/1211) by @JasonOA888
-- queuefs 去重 memory semantic parent enqueue,[PR #792](https://github.com/volcengine/OpenViking/pull/792) by @Protocol-zero-0
-- 为 decrypt 短明文场景补充回归测试,[PR #1223](https://github.com/volcengine/OpenViking/pull/1223) by @yc111233
-- 为写接口实现 request-scoped wait 机制,[PR #1212](https://github.com/volcengine/OpenViking/pull/1212) by @zhoujh01
-- 将 agfs 重组为 Rust 实现的 ragfs,[PR #1221](https://github.com/volcengine/OpenViking/pull/1221) by @MaojiaSheng
-- 修复 session `_wait_for_previous_archive_done` 无限挂起问题,[PR #1235](https://github.com/volcengine/OpenViking/pull/1235) by @yc111233
-- 修复 `#1238`、`#1242` 和 `#1232` 涉及的问题,[PR #1243](https://github.com/volcengine/OpenViking/pull/1243) by @MaojiaSheng
-- Memory 侧继续做性能优化,[PR #1159](https://github.com/volcengine/OpenViking/pull/1159) by @chenjw
+- **存储**:S3FS 新增 `disable_batch_delete` 选项兼容 OSS,改进 RAGFS 路径 scope 回退到 prefix filters。
+- **Session & Memory**:修复首条消息时缺失 session 的自动创建,解决 Memory V2 config 初始化顺序问题。
+- **Bot**:修复多用户 memory commit、响应语言处理,确保 `afterTurn` 以正确角色存储消息并跳过心跳条目。
+- **安全 & CI**:移除 settings.py 中泄露的 token,bot proxy 响应中清除内部错误细节,CI 优化为条件 OS 矩阵。
+- **开发者体验**:新增场景化 API 测试,queue status 中暴露 re-enqueue 计数便于调试。
-## v0.3.3 (2026-04-03)
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.4...v0.3.5)
-# OpenViking v0.3.3
+## v0.3.4 (2026-04-09)
-## Highlights
+### 版本亮点
-- **评测与写入能力继续扩展**:新增 RAG benchmark 评测框架 [PR #825](https://github.com/volcengine/OpenViking/pull/825),补充 OpenClaw 的 LoCoMo eval 脚本与说明 [PR #1152](https://github.com/volcengine/OpenViking/pull/1152),并新增内容写入接口 [PR #1151](https://github.com/volcengine/OpenViking/pull/1151)。
-- **OpenClaw 插件可用性显著增强**:补充架构文档与图示 [PR #1145](https://github.com/volcengine/OpenViking/pull/1145),安装器不再覆盖 `gateway.mode` [PR #1149](https://github.com/volcengine/OpenViking/pull/1149),新增端到端 healthcheck 工具 [PR #1180](https://github.com/volcengine/OpenViking/pull/1180),支持 bypass session patterns [PR #1194](https://github.com/volcengine/OpenViking/pull/1194),并在 OpenViking 故障时避免阻塞 OpenClaw [PR #1158](https://github.com/volcengine/OpenViking/pull/1158)。
-- **测试与 CI 覆盖大幅补强**:OpenClaw 插件新增大规模单测套件 [PR #1144](https://github.com/volcengine/OpenViking/pull/1144),补充 e2e 测试 [PR #1154](https://github.com/volcengine/OpenViking/pull/1154),新增 OpenClaw2OpenViking 集成测试与 CI 流水线 [PR #1168](https://github.com/volcengine/OpenViking/pull/1168)。
-- **会话、解析与导入链路更稳健**:支持创建 session 时指定 `session_id` [PR #1074](https://github.com/volcengine/OpenViking/pull/1074),CLI 聊天端点优先级与 `grep --exclude-uri/-x` 能力得到增强 [PR #1143](https://github.com/volcengine/OpenViking/pull/1143) [PR #1174](https://github.com/volcengine/OpenViking/pull/1174),目录导入 UX / 正确性与扫描 warning 契约也进一步改善 [PR #1197](https://github.com/volcengine/OpenViking/pull/1197) [PR #1199](https://github.com/volcengine/OpenViking/pull/1199)。
-- **稳定性与安全性继续加固**:修复任务 API ownership 泄露问题 [PR #1182](https://github.com/volcengine/OpenViking/pull/1182),统一 stale lock 处理并补充 ownership checks [PR #1171](https://github.com/volcengine/OpenViking/pull/1171),修复 ZIP 乱码 [PR #1173](https://github.com/volcengine/OpenViking/pull/1173)、embedder dimensions 透传 [PR #1183](https://github.com/volcengine/OpenViking/pull/1183)、语义 DAG 增量更新缺失 summary 场景 [PR #1177](https://github.com/volcengine/OpenViking/pull/1177) 等问题。
+- OpenClaw 插件默认配置调整(`recallPreferAbstract` 和 `ingestReplyAssist` 默认 `false`),新增 eval 脚本和 recall 查询清洗。
+- Memory 和会话运行时稳定性增强:request-scoped 写等待、PID lock 回收、孤儿 compressor 引用、async contention 修复。
+- 安全边界收紧:HTTP 资源导入 SSRF 防护、无 API key 时 trusted mode 仅允许 localhost、可配置 embedding circuit breaker。
+- 生态扩展:Volcengine Vector DB STS Token、MiniMax-M2.7 provider、Lua parser、Bot channel mention。
+- CI/Docker:发布时自动更新 `main` 并 Docker Hub push,Gemini optional dependency 纳入镜像。
-## Upgrade Notes
+### 升级说明
-- OpenClaw 插件安装器不再写入 `gateway.mode`。如果你之前依赖安装流程自动改写该配置,升级后需要改为显式管理,见 [PR #1149](https://github.com/volcengine/OpenViking/pull/1149)。
-- 如果你使用 `--with-bot` 进行安装或启动,失败时现在会直接返回错误码;依赖“失败但继续执行”行为的脚本需要同步调整,见 [PR #1175](https://github.com/volcengine/OpenViking/pull/1175)。
-- 如果你接入 OpenAI Dense Embedder,自定义维度参数现在会正确传入 `embed()`;此前依赖默认维度行为的调用方建议复核配置,见 [PR #1183](https://github.com/volcengine/OpenViking/pull/1183)。
-- `ov status` 现在会展示 embedding 与 rerank 模型使用情况,便于排障与环境核对,见 [PR #1191](https://github.com/volcengine/OpenViking/pull/1191)。
-- 检索侧曾尝试加入基于 tags metadata 的 cross-subtree retrieval [PR #1162](https://github.com/volcengine/OpenViking/pull/1162),但已在本版本窗口内回滚 [PR #1200](https://github.com/volcengine/OpenViking/pull/1200),因此不应将其视为 `v0.3.3` 的最终可用能力。
-- `litellm` 依赖范围更新为 `>=1.0.0,<1.83.1`,升级时建议同步检查锁文件与兼容性,见 [PR #1179](https://github.com/volcengine/OpenViking/pull/1179)。
+- OpenClaw `recallPreferAbstract` 和 `ingestReplyAssist` 现在默认 `false`,如需旧行为需显式配置。
+- HTTP 资源导入默认启用私网 SSRF 防护。
+- 无 API key 的 trusted mode 仅允许 localhost 访问。
+- 写接口引入 request-scoped wait,如有外部编排依赖旧时序需复核。
-## What's Changed
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.3...v0.3.4)
-### Benchmark, Eval, CLI, and Writing
+## v0.3.3 (2026-04-03)
-- 新增 RAG benchmark 系统评测框架,[PR #825](https://github.com/volcengine/OpenViking/pull/825) by @sponge225
-- 优化 CLI chat endpoint 配置优先级,[PR #1143](https://github.com/volcengine/OpenViking/pull/1143) by @ruansheng8
-- 新增 OpenClaw 的 LoCoMo eval 脚本与 README,[PR #1152](https://github.com/volcengine/OpenViking/pull/1152) by @yeshion23333
-- 新增内容写入接口,[PR #1151](https://github.com/volcengine/OpenViking/pull/1151) by @zhoujh01
-- `ov cli grep` 新增 `--exclude-uri` / `-x` 选项,[PR #1174](https://github.com/volcengine/OpenViking/pull/1174) by @heaoxiang-ai
-- 更新 rerank 配置文档,[PR #1138](https://github.com/volcengine/OpenViking/pull/1138) by @ousugo
+### 重点更新
-### OpenClaw Plugin, OpenCode Plugin, Bot, and Console
+- 新增 RAG benchmark 评测框架、OpenClaw LoCoMo eval 脚本、内容写入接口。
+- OpenClaw 插件:架构文档补充、安装器不再覆盖 `gateway.mode`、端到端 healthcheck 工具、bypass session patterns、OpenViking 故障隔离。
+- 测试覆盖:OpenClaw 插件单测、e2e 测试、oc2ov 集成测试与 CI。
+- Session 支持指定 `session_id` 创建;CLI 聊天端点优先级与 `grep --exclude-uri/-x` 增强。
+- 安全:任务 API ownership 泄露修复、stale lock 统一处理、ZIP 编码修复、embedder 维度透传。
-- 刷新 OpenClaw 插件架构文档并补充图示,[PR #1145](https://github.com/volcengine/OpenViking/pull/1145) by @qin-ctx
-- 修复 OpenClaw 插件安装器不应写入 `gateway.mode`,[PR #1149](https://github.com/volcengine/OpenViking/pull/1149) by @LinQiang391
-- 处理 OpenViking 故障时不再阻塞 OpenClaw,[PR #1158](https://github.com/volcengine/OpenViking/pull/1158) by @wlff123
-- 新增 OpenClaw 插件端到端 healthcheck 工具,[PR #1180](https://github.com/volcengine/OpenViking/pull/1180) by @wlff123
-- 新增 bypass session patterns 支持,[PR #1194](https://github.com/volcengine/OpenViking/pull/1194) by @Mijamind719
-- 修复 OpenCode 插件在当前 main 上恢复 stale commit state 的问题,[PR #1187](https://github.com/volcengine/OpenViking/pull/1187) by @13ernkastel
-- 新增 Bot 单通道(BotChannel)集成与 Werewolf demo,[PR #1196](https://github.com/volcengine/OpenViking/pull/1196) by @yeshion23333
-- console 支持 account user agentid,[PR #1198](https://github.com/volcengine/OpenViking/pull/1198) by @zhoujh01
+### 升级说明
-### Sessions, Retrieval, Parsing, and Resource Import
+- OpenClaw 安装器不再写入 `gateway.mode`,升级后需显式管理。
+- `--with-bot` 失败时返回错误码,依赖"失败但继续"行为的脚本需调整。
+- OpenAI Dense Embedder 自定义维度现正确传入 `embed()`。
+- 基于 tags metadata 的 cross-subtree retrieval 已在本版本窗口内回滚,非最终能力。
+- `litellm` 依赖更新为 `>=1.0.0,<1.83.1`。
-- 支持创建 session 时指定 `session_id`,[PR #1074](https://github.com/volcengine/OpenViking/pull/1074) by @likzn
-- 修复 reindex 时 memory `context_type` 的 URI 匹配逻辑,[PR #1155](https://github.com/volcengine/OpenViking/pull/1155) by @deepakdevp
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.2...v0.3.3)
## v0.3.2 (2026-04-01)
-## What's Changed
-* Chore/pr agent ark token billing
-* Fix HTTPX recognition issue with SOCKS5 proxy causing OpenViking crash
-* refactor(model): unify config-driven retry across VLM and embedding
-* fix(bot): import eval session time
-* docs(examples): retire legacy integration examples
-* fix(ci): skip filesystem tests on Windows due to FUSE compatibility
-* docs(docker): use latest image tag in examples
-* upload the new wechat group qrcode
-* Unify test directory in openclaw-plugin
-* docs(guides):Add concise OVPack guide in Chinese and English
-* docs(guides): reorganize observability documentation
-* fix(installer): fall back to official PyPI when mirror lags
-* feat(docker) add vikingbot and console
-* fix(vlm): rollback ResponseAPI to Chat Completions, keep tool calls
-* feat(openclaw-plugin): add session-pattern guard for ingest reply assist
-
-## v0.3.1 (2026-03-31)
+### 重点更新
-## What's Changed
-* feat(ast): add PHP tree-sitter support
-* feat(ci): add multi-platform API test support for 5 platforms
-* fix(ci): refresh uv.lock for docker release build
-* fix(openclaw-plugin): simplify install flow and harden helpers
-* fix(openclaw-plugin): preserve existing ov.conf on auto install
-* ci: build docker images natively per arch
-* Feature/memory opt
-* feat(storage): add auto language detection for semantic summary generation
-* feat(prompt) support configurable prompt template directories
-* feat: unify config-driven retry across VLM and embedding
-* feat(bot): import ov eval script
-* Revert "feat: unify config-driven retry across VLM and embedding"
-* fix(storage) Fix parent_uri compatibility with legacy records
-* fix(session): unify archive context abstracts
+- **Docker**:新增 VikingBot 和 Console 服务到 Docker 配置;示例更新为使用 latest 镜像标签。
+- **OpenClaw 插件**:新增 ingest reply assist 的 session-pattern guard;统一测试目录结构。
+- **VLM**:回滚 ResponseAPI 到 Chat Completions 同时保留 tool call 支持。
+- **稳定性**:修复 HTTPX SOCKS5 代理导致的崩溃;改进安装器 PyPI 镜像回退;Windows 上跳过 FUSE 不兼容的文件系统测试。
+- **文档**:新增中英文 OVPack 指南;重组可观测性文档;下线过时的集成示例。
-## v0.2.14 (2026-03-30)
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.3.1...v0.3.2)
-# OpenViking v0.2.14
+## v0.3.1 (2026-03-31)
-## Highlights
+### 重点更新
-- 多租户与身份管理进一步完善。CLI 已支持租户身份默认值与覆盖,文档新增多租户使用指南,memory 也支持仅按 agent 维度隔离的 `agent-only` scope。
-- 解析与导入链路更完整。图片解析新增 OCR 文本提取,目录导入识别 `.cc` 文件,重复标题导致的文件名冲突得到修复,HTTP 上传链路改为更稳妥的 upload id 流程。
-- OpenClaw 插件显著增强。安装器与升级流程统一,默认按最新 Git tag 安装,session API 与 context pipeline 做了统一重构,并补齐了 Windows、compaction、compact result mapping、子进程重拉起等多处兼容性与稳定性问题。
-- Bot 与 Feishu 集成可用性继续提升。修复了 bot proxy 未鉴权问题,改进了 Moonshot 请求兼容性,并升级了 Feishu interactive card 的 markdown 展示体验。
-- 存储与运行时稳定性持续提升。包括 queuefs embedding tracker 加固、vector store 移除 `parent_uri`、Docker doctor 检查对齐,以及更细粒度的 eval token 指标。
+- **语言支持**:新增 PHP tree-sitter AST 解析。
+- **存储**:语义摘要生成引入自动语言检测;修复 legacy 记录的 parent URI 兼容性。
+- **CI**:API 测试扩展到 5 个平台;切换为按架构原生构建 Docker 镜像;刷新 uv.lock 用于发布构建。
+- **配置**:新增可配置 prompt 模板目录;统一 session 管理中的 archive context 处理。
+- **OpenClaw 插件**:简化安装流程、加固辅助工具、自动安装时保留已有 `ov.conf`。
+- **Memory**:应用 memory 优化改进。
-## Upgrade Notes
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.14...v0.3.1)
-- Bot proxy 接口 `/bot/v1/chat` 与 `/bot/v1/chat/stream` 已补齐鉴权,依赖未鉴权访问的调用方需要同步调整,见 [#996](https://github.com/volcengine/OpenViking/pull/996)。
-- 裸 HTTP 导入本地文件/目录时,推荐按 `temp_upload -> temp_file_id` 的方式接入上传链路,见 [#1012](https://github.com/volcengine/OpenViking/pull/1012)。
-- OpenClaw 插件的 compaction delegation 修复要求 `openclaw >= v2026.3.22`,见 [#1000](https://github.com/volcengine/OpenViking/pull/1000)。
-- OpenClaw 插件安装器现在默认跟随仓库最新 Git tag 安装,如需固定版本可显式指定,见 [#1050](https://github.com/volcengine/OpenViking/pull/1050)。
+## v0.2.14 (2026-03-30)
-## What's Changed
+### 重点更新
-### Multi-tenant, Memory, and Identity
+- 多租户与身份管理:CLI 租户身份默认值与覆盖、`agent-only` memory scope、多租户使用指南。
+- 解析与导入:图片 OCR 文本提取、`.cc` 文件识别、重复标题文件名冲突修复、upload-id 方式 HTTP 上传。
+- OpenClaw 插件:统一安装器/升级流程、默认按最新 Git tag 安装、session API 与 context pipeline 重构、Windows/compaction/子进程兼容性修复。
+- Bot 与 Feishu:proxy 鉴权修复、Moonshot 兼容性改进、Feishu interactive card markdown 升级。
+- 存储与运行时:queuefs embedding tracker 加固、vector store `parent_uri` 移除、Docker doctor 对齐、eval token 指标。
-- 支持 CLI 租户身份默认值与覆盖,[#1019](https://github.com/volcengine/OpenViking/pull/1019) by @zhoujh01
-- 支持仅按 agent 隔离的 agent memory scope,[#954](https://github.com/volcengine/OpenViking/pull/954) by @liberion1994
-- 新增多租户使用指南文档,[#1029](https://github.com/volcengine/OpenViking/pull/1029) by @qin-ctx
-- 在 quickstart 和 basic usage 中补充 `user_key` 与 `root_key` 的区别说明,[#1077](https://github.com/volcengine/OpenViking/pull/1077) by @r266-tech
-- 示例中支持将 recalled memories 标记为 used,[#1079](https://github.com/volcengine/OpenViking/pull/1079) by @0xble
+### 升级说明
-### Parsing and Resource Import
+- Bot proxy 接口 `/bot/v1/chat` 和 `/bot/v1/chat/stream` 已补齐鉴权。
+- HTTP 导入推荐按 `temp_upload → temp_file_id` 方式接入。
+- OpenClaw 插件 compaction delegation 要求 `openclaw >= v2026.3.22`。
+- OpenClaw 安装器默认跟随最新 Git tag,如需固定版本可显式指定。
-- 新增图片解析 OCR 文本提取能力,[#942](https://github.com/volcengine/OpenViking/pull/942) by @mvanhorn
-- 修复重复标题导致的合并文件名冲突,[#1005](https://github.com/volcengine/OpenViking/pull/1005) by @deepakdevp
-- 目录导入时识别 `.cc` 文件,[#1008](https://github.com/volcengine/OpenViking/pull/1008) by @qin-ctx
-- 增加对非空 S3 目录 marker 的兼容性,[#997](https://github.com/volcengine/OpenViking/pull/997) by @zhoujh01
-- HTTP 上传链路从临时路径切换为 upload id,[#1012](https://github.com/volcengine/OpenViking/pull/1012) by @qin-ctx
-- 修复从目录上传时临时归档文件丢失 `.zip` 后缀的问题,[#1021](https://github.com/volcengine/OpenViking/pull/1021) by @Shawn-cf-o
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.13...v0.2.14)
-### OpenClaw Plugin and Installer
+## v0.2.13 (2026-03-26)
-- 回滚 duplicate registration guard 的加固改动,[#995](https://github.com/volcengine/OpenViking/pull/995) by @qin-ptr
-- Windows 下 mapped session ID 做安全清洗,[#998](https://github.com/volcengine/OpenViking/pull/998) by @qin-ctx
-- 使用 plugin-sdk exports 实现 compaction delegation,修复 `#833`,要求 `openclaw >= v2026.3.22`,[#1000](https://github.com/volcengine/OpenViking/pull/1000) by @jcp0578
-- 统一 installer 升级流程,[#1020](https://github.com/volcengine/OpenViking/pull/1020) by @LinQiang391
-- 默认按最新 Git tag 安装 OpenClaw 插件,[#1050](https://github.com/volcengine/OpenViking/pull/1050) by @LinQiang391
-- 统一 session APIs,并重构 OpenClaw context pipeline 以提升一致性、可维护性和测试覆盖,[#1040](https://github.com/volcengine/OpenViking/pull/1040) by @wlff123
-- 将 openclaw-plugin 的 `DEFAULT_COMMIT_TOKEN_THRESHOLD` 调整为 `20000`,[#1052](https://github.com/volcengine/OpenViking/pull/1052) by @wlff123
+### 重点更新
-## v0.2.13 (2026-03-26)
+- **测试**:新增核心工具的全面单元测试;改进 API 测试基础设施支持双模式 CI。
+- **平台**:修复 Windows engine wheel 运行时打包。
+- **VLM**:LiteLLM thinking 参数限定为 DashScope provider。
+- **OpenClaw 插件**:加固重复注册 guard。
+- **文档**:新增基础用法示例和中文文档。
-## What's Changed
-* test: add comprehensive unit tests for core utilities
-* fix(vlm): scope LiteLLM thinking param to DashScope providers only
-* Api test:improve API test infrastructure with dual-mode CI
-* docs: Add basic usage example and Chinese documentation for examples
-* Fix Windows engine wheel runtime packaging
-* fix(openclaw-plugin): harden duplicate registration guard
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.12...v0.2.13)
## v0.2.12 (2026-03-25)
-## What's Changed
-* Use uv sync --locked in Dockerfile
-* fix(server): handle CancelledError during shutdown paths
-* fix(bot):rollback config
+此补丁版本通过正确处理 `CancelledError` 稳定了服务器 shutdown 序列,回滚了一个 bot 配置回退,并通过切换到 `uv sync --locked` 加强 Docker 构建的依赖一致性。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.11...v0.2.12)
## v0.2.11 (2026-03-25)
-# OpenViking v0.2.11
-
-## 版本亮点
-
-- **模型与检索生态继续扩展**:新增 MiniMax embedding [#624](https://github.com/volcengine/OpenViking/pull/624)、Azure OpenAI embedding/VLM [#808](https://github.com/volcengine/OpenViking/pull/808)、GeminiDenseEmbedder [#751](https://github.com/volcengine/OpenViking/pull/751)、LiteLLM embedding [#853](https://github.com/volcengine/OpenViking/pull/853) 与 rerank [#888](https://github.com/volcengine/OpenViking/pull/888),并补充 OpenAI-compatible rerank [#785](https://github.com/volcengine/OpenViking/pull/785) 与 Tavily 搜索后端 [#788](https://github.com/volcengine/OpenViking/pull/788)。
-- **内容接入链路更完整**:音频资源现在支持 Whisper ASR 解析 [#805](https://github.com/volcengine/OpenViking/pull/805),云文档场景新增飞书/Lark 解析器 [#831](https://github.com/volcengine/OpenViking/pull/831),文件向量化策略变为可配置 [#858](https://github.com/volcengine/OpenViking/pull/858),搜索结果还新增了 provenance 元数据 [#852](https://github.com/volcengine/OpenViking/pull/852)。
-- **服务端运维能力明显补齐**:新增 `ov reindex` [#795](https://github.com/volcengine/OpenViking/pull/795)、`ov doctor` [#851](https://github.com/volcengine/OpenViking/pull/851)、Prometheus exporter [#806](https://github.com/volcengine/OpenViking/pull/806)、内存健康统计 API [#706](https://github.com/volcengine/OpenViking/pull/706)、可信租户头模式 [#868](https://github.com/volcengine/OpenViking/pull/868) 与 Helm Chart [#800](https://github.com/volcengine/OpenViking/pull/800)。
-- **多租户与安全能力增强**:新增多租户文件加密 [#828](https://github.com/volcengine/OpenViking/pull/828) 和文档加密能力 [#893](https://github.com/volcengine/OpenViking/pull/893),修复租户上下文在 observer 与 reindex 流程中的透传问题 [#807](https://github.com/volcengine/OpenViking/pull/807) [#820](https://github.com/volcengine/OpenViking/pull/820),并修复 ZIP Slip 风险 [#879](https://github.com/volcengine/OpenViking/pull/879) 与 trusted auth 模式下 API key 校验缺失 [#924](https://github.com/volcengine/OpenViking/pull/924)。
-- **稳定性与工程体验持续加固**:向量检索 NaN/Inf 分数处理在结果端 [#824](https://github.com/volcengine/OpenViking/pull/824) 和源头 [#882](https://github.com/volcengine/OpenViking/pull/882) 双重兜底,会话异步提交与并发提交问题被修复 [#819](https://github.com/volcengine/OpenViking/pull/819) [#783](https://github.com/volcengine/OpenViking/pull/783) [#900](https://github.com/volcengine/OpenViking/pull/900),Windows stale lock 与 TUI 输入问题持续修复 [#790](https://github.com/volcengine/OpenViking/pull/790) [#798](https://github.com/volcengine/OpenViking/pull/798) [#854](https://github.com/volcengine/OpenViking/pull/854),同时补上了代理兼容 [#957](https://github.com/volcengine/OpenViking/pull/957) 与 API 重试风暴保护 [#772](https://github.com/volcengine/OpenViking/pull/772)。
-
-## 升级提示
-
-- 如果你使用 `litellm` 集成,请关注这一版本中的安全策略调整:先临时硬禁用 [#937](https://github.com/volcengine/OpenViking/pull/937),后恢复为仅允许 `<1.82.6` 的版本范围 [#966](https://github.com/volcengine/OpenViking/pull/966)。建议显式锁定依赖版本。
-- 如果你启用 trusted auth 模式,需要同时配置服务端 API key,相关校验已在 [#924](https://github.com/volcengine/OpenViking/pull/924) 中强制执行。
-- Helm 默认配置已切换为更适合 Volcengine 场景的默认值,并补齐缺失字段 [#822](https://github.com/volcengine/OpenViking/pull/822)。升级 chart 时建议重新审阅 values 配置。
-- Windows 用户建议关注 stale PID / RocksDB LOCK 文件处理增强 [#790](https://github.com/volcengine/OpenViking/pull/790) [#798](https://github.com/volcengine/OpenViking/pull/798) [#854](https://github.com/volcengine/OpenViking/pull/854)。
-
-### 模型、Embedding、Rerank 与检索生态
-
-- 新增 MiniMax embedding provider,支持通过官方 HTTP API 接入 MiniMax 向量模型。[PR #624](https://github.com/volcengine/OpenViking/pull/624)
-- 新增 OpenAI-compatible rerank provider。[PR #785](https://github.com/volcengine/OpenViking/pull/785)
-- 新增 Azure OpenAI 对 embedding 与 VLM 的支持。[PR #808](https://github.com/volcengine/OpenViking/pull/808)
-- 新增 GeminiDenseEmbedder 文本向量模型提供方。[PR #751](https://github.com/volcengine/OpenViking/pull/751)
-- 新增 Tavily 作为可配置的 web search backend。[PR #788](https://github.com/volcengine/OpenViking/pull/788)
-- 修复 LiteLLM 下 `zai/` 模型前缀处理,避免重复拼接 `zhipu` 前缀。[PR #789](https://github.com/volcengine/OpenViking/pull/789)
-- 修复 Gemini embedder 相关问题。[PR #841](https://github.com/volcengine/OpenViking/pull/841)
-- 新增 LiteLLM embedding provider。[PR #853](https://github.com/volcengine/OpenViking/pull/853)
-- 新增默认向量索引名可配置能力。[PR #861](https://github.com/volcengine/OpenViking/pull/861)
-- 新增 LiteLLM rerank provider。[PR #888](https://github.com/volcengine/OpenViking/pull/888)
-- 修复 Ollama embedding provider 的维度解析问题。[PR #915](https://github.com/volcengine/OpenViking/pull/915)
-- 为 Jina 代码模型补充 code-specific task 默认值。[PR #914](https://github.com/volcengine/OpenViking/pull/914)
-- 在 embedder 与 vectordb 维度不匹配时给出警告提示。[PR #930](https://github.com/volcengine/OpenViking/pull/930)
-- 为 Jina embedding provider 增加代码模型维度与更可操作的 422 错误提示。[PR #928](https://github.com/volcengine/OpenViking/pull/928)
-- 搜索结果新增 provenance 元数据,方便追踪召回来源。[PR #852](https://github.com/volcengine/OpenViking/pull/852)
-- 修复 recall limit 逻辑。[PR #821](https://github.com/volcengine/OpenViking/pull/821)
-- 在结果序列化前钳制向量检索中的 `inf`/`nan` 分数,避免 JSON 失败。[PR #824](https://github.com/volcengine/OpenViking/pull/824)
-- 在相似度分数源头钳制 `inf`/`nan`,进一步避免 JSON crash。[PR #882](https://github.com/volcengine/OpenViking/pull/882)
-
-### 解析、导入与内容处理
-
-- 新增基于 Whisper 的音频解析与 ASR 集成。[PR #805](https://github.com/volcengine/OpenViking/pull/805)
-- 修复 PDF 书签目标页为整数索引时的解析问题。[PR #794](https://github.com/volcengine/OpenViking/pull/794)
-- 新增飞书/Lark 云文档解析器。[PR #831](https://github.com/volcengine/OpenViking/pull/831)
-- 修复 MarkdownParser 的字符数限制处理。[PR #826](https://github.com/volcengine/OpenViking/pull/826)
-- 支持可配置的文件向量化策略。[PR #858](https://github.com/volcengine/OpenViking/pull/858)
+### 版本亮点
-## v0.2.10 (2026-03-24)
+- 模型与检索生态扩展:MiniMax embedding、Azure OpenAI embedding/VLM、GeminiDenseEmbedder、LiteLLM embedding 和 rerank、OpenAI-compatible rerank、Tavily 搜索后端。
+- 内容接入:Whisper ASR 音频解析、飞书/Lark 云文档解析器、可配置文件向量化策略、搜索结果 provenance 元数据。
+- 服务端运维:`ov reindex`、`ov doctor`、Prometheus exporter、内存健康统计 API、可信租户头模式、Helm Chart。
+- 多租户与安全:多租户文件加密和文档加密、租户上下文透传修复、ZIP Slip 修复、trusted auth API key 强制校验。
+- 稳定性:向量检索 NaN/Inf 分数钳制、异步/并发 session commit 修复、Windows stale lock 和 TUI 修复、代理兼容、API 重试风暴保护。
-# LiteLLM 安全热修复 Release Note
-
-更新时间:2026-03-24
+### 升级提示
-## 背景
+- `litellm` 安全策略调整:先临时禁用,后恢复为 `<1.82.6` 版本范围。建议显式锁定依赖版本。
+- trusted auth 模式需同时配置服务端 API key。
+- Helm 默认配置切换为 Volcengine 场景默认值,升级时建议重新审阅 values。
-由于上游依赖 `LiteLLM` 出现公开供应链安全事件,OpenViking 在本次热修复中临时禁用所有 LiteLLM 相关入口,以避免继续安装或运行到受影响依赖。
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.10...v0.2.11)
-## 变更内容
+## v0.2.10 (2026-03-24)
-- 移除根依赖中的 `litellm`
-- 移除根 `uv.lock` 中的 `litellm`
-- 禁用 LiteLLM 相关的 VLM provider 入口
-- 禁用 bot 侧 LiteLLM provider 和图片工具入口
-- 增加 LiteLLM 已禁用的回归测试
+### LiteLLM 安全热修复
-## 建议操作
+由于上游依赖 `LiteLLM` 出现公开供应链安全事件,本次热修复临时禁用所有 LiteLLM 相关入口。
-建议用户立即执行以下动作:
+### 建议操作
1. 检查运行环境中是否安装 `litellm`
2. 卸载可疑版本并重建虚拟环境、容器镜像或发布产物
3. 对近期安装过可疑版本的机器轮换 API Key 和相关凭证
4. 升级到本热修复版本
-可用命令:
+LiteLLM 相关能力会暂时不可用,直到上游给出可信的修复版本和完整事故说明。
-## 兼容性说明
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.9...v0.2.10)
-这是一个以止损为目标的防御性热修复版本。LiteLLM 相关能力会暂时不可用,直到上游给出可信的修复版本和完整事故说明。
+## v0.2.9 (2026-03-19)
-## 参考链接
+此版本聚焦于稳定性和开发者体验改进。关键修复包括:通过在 account backend 间共享单一 adapter 解决 RocksDB 锁竞争、恢复之前合并中丢失的插件 bug fix、改善 vector store 增量更新。新功能包括 bot 调试模式和 `/remember` 命令、semantic pipeline 中基于 summary 的文件 embedding、CI 中全面的 PR-Agent 评审规则。文档新增 Docker Compose 和 Mac 端口转发指引。
-- 上游 issue: <https://github.com/BerriAI/litellm/issues/24512>
-- PyPI 项目页: <https://pypi.org/project/litellm/>
-- GitHub Security: <https://github.com/BerriAI/litellm/security>
-- OpenViking 热修分支: <https://github.com/volcengine/OpenViking/tree/hotfix/v0.2.10-disable-litellm>
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.8...v0.2.9)
-## v0.2.9 (2026-03-19)
+## v0.2.8 (2026-03-19)
-## What's Changed
-* fix(resource): enforce agent-level watch task isolation
-* feat(embedder): use summary for file embedding in semantic pipeline
-* Fix/bot readme
-* Fix/increment update dir vector store
-* fix(plugin): restore bug fixes from #681 and #688 lost during #662 merge
-* docs: add docker compose instructions and mac port forwarding tip to …
-* Update docs
-* feat(ci): add comprehensive Qodo PR-Agent review rules
-* [feat](bot):Add mode config, add debug mode, add /remember cmd
-* fix(vectordb): share single adapter across account backends to prevent RocksDB lock contention
+### 重点更新
-## v0.2.8 (2026-03-19)
+- OpenClaw 插件升级到 2.0(context engine),新增 OpenCode memory plugin,多智能体 memory isolation 基于 `agentId`。
+- Memory 冷热分层 archival 和 hotness scoring、长记忆 chunked vectorization、`used()` 使用追踪接口。
+- 分层检索集成 rerank、RetrievalObserver 检索质量观测。
+- 资源 watch scheduling、reindex endpoint、legacy `.doc`/`.xls` 解析支持、path locking 和 crash recovery。
+- 请求级 trace metrics、memory extract telemetry breakdown、OpenAI VLM streaming、`<think>` 标签自动清理。
+- 跨平台修复(Windows zip、Rust CLI)、AGFS Makefile 重构、CPU variant vectordb engine、Python 3.14 wheel 支持。
-# OpenViking v0.2.8 发布公告
-
-## 本次更新亮点
-
-### 1. 插件生态继续升级,OpenClaw / OpenCode 集成更完整
-- `openclaw-plugin` 升级到 **2.0**,从 memory plugin 进一步演进为 **context engine**。
-- 新增并完善 **OpenCode memory plugin example**,补充 attribution 与后续插件更新。
-- 支持 **多智能体 memory isolation**,可基于 hook context 中的 `agentId` 做记忆隔离。
-- 修复插件自动 recall 可能导致 agent 长时间挂起的问题,并补强 legacy commit 兼容。
-- MCP 查询服务补充 `api_key` 支持与默认配置能力,降低接入成本。
-
-### 2. Session / Memory 链路增强,长期记忆能力更稳
-- 新增 **memory cold-storage archival**,通过 hotness scoring 管理长期记忆冷热分层。
-- 新增长记忆 **chunked vectorization**,改善超长内容的向量化处理能力。
-- 增加 `used()` 接口,用于上下文 / skill 使用追踪。
-- 修复 async commit 过程中的上下文透传、批次内重复写入、非标准 LLM 响应处理等问题。
-- Bot 场景下改用 `commit_async()`,减少阻塞风险,提升长会话稳定性。
-
-### 3. 检索与 Embedding 能力持续增强
-- 分层检索中正式集成 **rerank**,并修复无 rerank 场景下的检索可用性问题。
-- 新增 **RetrievalObserver**,可用于检索质量指标观测,并已接入 `ov observer`。
-- Embedding 侧新增:
-- 修复 CJK token 估算偏低等问题,提升多语言场景下的稳定性。
-
-### 4. 资源、存储与解析链路更完善
-- 新增 **resource watch scheduling** 与状态跟踪能力,资源同步流程更可控。
-- 新增 **reindex endpoint**,支持内容手动修改后的重新 embedding。
-- 解析能力新增对 **legacy `.doc` / `.xls`** 格式的支持。
-- 修复从 URL 导入文件时文件名与扩展名丢失的问题。
-- 存储层新增 **path locking** 与选择性 crash recovery,进一步提升写入安全性。
-- 修复 tenant API 的隐式 root fallback、文件系统接口路径处理、工作目录 `~` 展开等问题。
-
-### 5. VLM、Trace 与可观测性能力加强
-- 新增 **request-level trace metrics** 与对应 API 支持。
-- 新增 **memory extract telemetry breakdown**,帮助更细粒度地分析记忆提取过程。
-- OpenAI VLM 支持 **streaming response handling**。
-- 补充 `max_tokens` 参数以避免 vLLM 拒绝请求,并支持 OpenAI-compatible VLM 的自定义 HTTP headers。
-- 自动清理模型输出中的 `<think>` 标签,减少推理内容污染存储结果。
-
-### 6. 工程兼容性与交付体验继续改进
-- 修复 Windows zip 路径、代码仓库索引、Rust CLI 版本等跨平台问题。
-- `agfs` Makefile 完成跨平台兼容性重构。
-- Vectordb engine 支持按 **CPU variant** 拆分。
-- Docker 构建链路持续修复,补齐 `build_support/` 拷贝逻辑。
-- Release workflow 支持发布 **Python 3.14 wheels**。
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.6...v0.2.8)
## v0.2.6 (2026-03-11)
-# OpenViking v0.2.6 发布公告
+### 重点更新
-## 重点更新
+- CLI 体验:`ov chat` 基于 `rustyline` 行编辑、Markdown 渲染、聊天历史。
+- 异步能力:session commit `wait` 参数、可配置 worker count。
+- 新增 OpenViking Console Web 控制台,方便调试和 API 探索。
+- Bot 增强:eval 能力、`add-resource` 工具、飞书进度通知。
+- OpenClaw memory plugin 大幅升级:npm 安装、统一安装器、稳定性修复。
+- 平台支持:Linux ARM、Windows UTF-8 BOM 修复、CI runner OS 固定。
-### 1. CLI 与对话体验进一步升级
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.5...v0.2.6)
-- `ov chat` 现在基于 `rustyline` 提供更完整的行编辑体验,终端交互更自然,不再出现常见的方向键控制字符问题。
-- 新增 Markdown 渲染能力,终端中的回答展示更清晰,代码块、列表等内容可读性更好。
-- 支持聊天历史记录,同时提供关闭格式化和关闭历史记录的选项,便于在不同终端环境中按需使用。
+## v0.2.5 (2026-03-06)
-### 2. 服务端异步能力增强,长任务不再轻易阻塞
+### 重点更新
-- Session commit 新增异步提交能力,并支持通过 `wait` 参数控制是否同步等待结果。
-- 当选择后台执行时,OpenViking 现在会返回可追踪的任务信息,调用方可以通过任务接口查询状态、结果或错误。
-- 服务端新增可配置 worker count,进一步缓解单 worker 场景下长任务阻塞请求的问题。
+- **PDF & 解析**:基于字体的标题检测和书签提取为结构化 markdown 标题;`add_resource` 支持索引控制并重构 embedding 逻辑,正确处理 ZIP 容器格式。
+- **Session & Memory**:`add_message()` 新增 `parts` 参数支持;memory 抽取后为父目录触发语义索引。
+- **URI 处理**:短格式 `VikingURI` 支持、CLI 中 `git@` SSH URL 格式、GitHub `tree/<ref>` URL 代码仓库导入。
+- **Bot & 集成**:VikingBot 重构包含新评测模块、飞书多用户和 channel 增强、OpenAPI 标准化;Telegram Claude 崩溃修复。
+- **基础设施**:`agfs` 新增 ripgrep 加速的 grep 和 async grep,可选 binding client 模式;使用 Doubao 模型的自动 PR 评审工作流和严重度分级。
+- **安装**:curl 方式安装在 Ubuntu/Debian 上不再触发系统保护错误;修复 `uv pip install -e .` 的 Rust 编译。
-### 3. 新增 Console,并持续增强 Bot 与资源操作能力
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.3...v0.2.5)
-- 本版本新增独立的 OpenViking Console,提供更直观的 Web 控制台入口,方便调试、调用和查看接口结果。
-- Bot 能力继续增强,新增 eval 能力,开放 `add-resource` 工具,并支持飞书进度通知等扩展能力。
-- 资源导入支持 `preserve_structure` 选项,扫描目录时可以保留原始目录层级,适合更复杂的知识组织方式。
-- 同时修复了 grep、glob、`add-resource` 等场景下的一些响应问题,提升日常使用稳定性。
+## v0.2.3 (2026-03-03)
-### 4. openclaw memory plugin 能力和稳定性大幅完善
+### Breaking Change
-- openclaw memory plugin 在这一版中获得了较大幅度增强,补充了更完整的插件能力与相关示例。
-- 插件安装链路进一步完善,现已支持通过 npm 包方式安装 setup-helper,部署和升级都更直接。
-- 修复了本地运行、端口管理、配置保留、缺失文件下载、配置覆盖等一系列影响安装和使用体验的问题。
-- Memory consolidation 与 skill tool memory 相关问题也得到修复,进一步提升记忆链路的稳定性。
-- Vikingbot 作为 OpenViking 可选依赖的集成方式也做了完善,降低了插件和 bot 协同使用时的接入门槛。
+升级后,历史版本生成的 datasets/indexes 与新版本不兼容,无法直接复用。升级后需要全量重建数据集以避免检索异常、过滤结果不一致或运行时错误。停止服务,删除 workspace 目录(`rm -rf ./your-openviking-workspace`),然后用 `openviking-server` 重启。
-### 5. 安装、跨平台与工程质量继续补强
+此版本提供 CLI 优化,包括 `glob -n` 标志支持和 `cmd echo`,以及中英文 README 更新。
-- 新增 Linux ARM 支持,进一步扩展了 OpenViking 的部署平台范围。
-- 修复了 Windows 下 UTF-8 BOM 配置文件的兼容问题,减少配置读取失败的情况。
-- 修复了 `install.sh`、`ov.conf.example`、setup 失败退出等问题,并补充了 openclaw memory plugin 的 npm 包安装路径,提升首次安装和配置过程的成功率。
-- CI 侧将 GitHub Actions runner 固定到明确 OS 版本,减少环境漂移带来的构建与发布不确定性。
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.2...v0.2.3)
-## 总结
+## v0.2.2 (2026-03-03)
-## What's Changed
-* feat: improve ov chat UX with rustyline and markdown rendering
-* fix: grep for binding-client
-* feat(server): add configurable worker count to prevent single-worker blocking
-* feat(sessions): add async commit support with wait parameter
+### Breaking Change
-## v0.2.5 (2026-03-06)
+升级前请先停止 VikingDB Server 并清除 workspace 目录。旧版本的索引与此版本不向前兼容。
-## What's Changed
-* docs: use openviking-server to launch server
-* fix: Session.add_message() support parts parameter
-* feat: support GitHub tree/&lt;ref&gt; URL for code repository import
-* fix: improve ISO datetime parsing
-* feat(pdf): extract bookmarks as markdown headings for hierarchical parsing
-* feat: add index control to add_resource and refactor embedding logic
-* fix: support short-format URIs in VikingURI and VikingFS access control
-* fix: handle None data in skill_processor._parse_skill()
-* fix: fix add-resource
-* fix: treat only .zip as archive; avoid unzipping ZIP-based container …
-* fix(cli): support git@ SSH URL format in add-resource
-* feat(pdf): add font-based heading detection and refactor PDF/Markdown parsing
-* 支持通过curl方式安装部署openclaw+openviking插件
-* Feature/vikingbot_opt: OpenAPI interface standardization;Feishu multi-user experience; observability enhancements; configuration system modernization.
-* docs: pip upgrade suggestion
-* feat: define a system path for future deployment
-* chore: downgrade golang version limit, update vlm version to seed 2.0
-* [fix]: 修复通过curl命令安装场下下ubuntu/debian等系统触发系统保护无法安装的openviking的问题
-* fix(session): trigger semantic indexing for parent directory after memory extraction
-* feat(bot):Refactoring, New Evaluation Module, Feishu channel Opt & Feature Enhancements
-* chore: agfs-client默认切到binding-client模式
-* feat(agfs): add ripgrep-based grep acceleration and fix vimgrep line parser
-* [WIP] ci: add automated PR review workflow using Doubao model
-* fix(ci): use OPENAI_KEY instead of OPENAI.KEY in pr-review workflow
-* build(deps): bump actions/download-artifact from 7 to 8 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 6 to 7 by @dependabot[bot] in
-* build(deps): bump docker/login-action from 3 to 4 by @dependabot[bot] in
-* build(deps): bump docker/setup-qemu-action from 3 to 4 by @dependabot[bot] in
-* build(deps): bump docker/setup-buildx-action from 3 to 4 by @dependabot[bot] in
-* fix(packaging): sdist 排除运行时二进制
-* ci: enhance PR review with severity classification and checklist
-* enable injection for Collection and CollectionAdaptor
-* chore: 编译子命令失败报错, golang版本最低要求1.22+
-* feat(viking_fs) support async grep
-* OpenViking Plugin Exception Handling & Fixing
-* feat(agfs): make binding client optional, add server bootstrap, tune logging and CI
-* fix: rust compile in uv pip install -e .
-* [wip]Enhance OpenViking Status Checks in the OpenClaw Plugin
-* FIX: fixes multiple issues in the OpenViking chat functionality and unifies session ID generation logic between Python and Rust CLI implementations.
-* chore: 增加Makefile,方便build
-* fix(bot): fix Telegram channel init crash and empty content for Claude
-* fix: suport uv pip install openviking[bot]
-* fix: agfs-client默认使用http-client
-* fix(agfs): fix agfs binding-client import error
+此版本新增 C# AST 提取器支持代码解析,修复多租户过滤,规范 OpenViking memory target paths,改进 `git@` SSH URL 的 git 仓库检测。`agfs` 依赖的 lib/bin 现在预编译提供,安装时无需构建步骤。文档新增千问模型使用说明。
-## v0.2.3 (2026-03-03)
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.2.1...v0.2.2)
-## Breaking Change
-After upgrading, datasets/indexes generated by historical versions are not compatible with the new version and cannot be reused directly. Please rebuild the datasets after upgrading (a full rebuild is recommended) to avoid retrieval anomalies, inconsistent filtering results, or runtime errors.
-Stop the service -> rm -rf ./your-openviking-workspace -> restart the service with the openviking-server command.
+## v0.2.1 (2026-02-28)
-## What's Changed
-* Feat: CLI optimization
-* Update README.md
-* Update README_CN.md
-* feat: support glob -n and cmd echo
-* fix: fix release
+### 重点更新
-## v0.2.2 (2026-03-03)
+- **多租户**:API 层多租户基础能力,支持多用户/团队隔离使用。
+- **云原生**:云原生 VikingDB 支持,完善云端部署文档和 Docker CI。
+- **OpenClaw/OpenCode**:官方 `openclaw-openviking-plugin` 安装、`opencode` 插件引入。
+- **存储**:向量数据库接口重构、AGFS binding client、AST 代码骨架提取、私有 GitLab 域名支持。
+- **CLI**:`ov` 命令封装、`add-resource` 增强、`ovcli.conf` timeout 支持、`--version` 参数。
-## Breaking Change
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.18...v0.2.1)
-Warning: This Release includes Breaking Chage! Before upgrading, you should stop VikingDB Server and clear workspace dir first.
+## [email protected] (2026-02-27)
-## What's Changed
-* fix ci
-* 在readme补充千问使用方法
-* feat(parse): Add C# AST extractor support
-* chore: bump CLI version to 0.2.1
-* fix: normalize OpenViking memory target paths
-* fix: fix filter when multi tenants
-* docs/update_wechat
-* Fix retriever
-* fix(git): support git@ SSH URLs with regression-safe repo detection
-* fix(agfs): 预编译agfs所依赖的lib/bin,无需安装时构建
+更新的 CLI 二进制发布,跨平台支持 macOS 和 Linux,与 v0.1.18 功能集对齐,包含 Rust 实现和扩展的文件解析器能力。
-## v0.2.1 (2026-02-28)
+[完整变更记录](https://github.com/volcengine/OpenViking/releases/tag/cli%400.2.0)
-## 1. Core Capability Upgrades: Multi-tenancy, Cloud-Native & OpenClaw/OpenCode Adaptation
-- **Multi-tenancy**: Implemented foundational multi-tenancy support at the API layer (#260, #283), laying the groundwork for isolated usage across multiple users/teams.
-- **Cloud-Native Support**: Added support for cloud-native VikingDB (#279), improved cloud deployment documentation and Docker CI workflows (#320), expanding from local deployment to cloud-edge integration.
-- **OpenClaw Integration**: Added official installation for the `openclaw-openviking-plugin` (#307), strengthening integration with OpenClaw.
-- **OpenCode Support**: Introduced the `opencode` plugin and updated documentation (#351), extending capabilities for code-related scenarios.
-
-## 2. Deep Optimization of the Database Storage Foundation
-- **Architecture Refactor**: Refactored the vector database interface (#327) and removed the `with_vector` parameter from query APIs (#338) to simplify the interface design.
-- **Performance Optimizations**:
-- **BugFixes**: Resolved missing Volcengine URI prefixes for VikingDB, invalidated `is_leaf` filters (#294), and fixed vector storage lock contention during fast restarts (#343).
-- **AGFS Enhancements**: Added AGFS binding client (#304), fixed AGFS SDK installation/import issues (#337, #355), and improved filesystem integration.
-- **Code Scenario Improvements**: Added AST-based code skeleton extraction mode (#334), supported private GitLab domains for code repositories (#285), and optimized GitHub ZIP download (#267).
-
-## 3. Improved CLI Toolchain (Greatly Enhanced Usability)
-- Added `ov` command wrapper (#325) and fixed bugs in the CLI wrapper, repo URI handling, and `find` command (#336, #339, #347).
-- Enhanced `add-resource` functionality with unit tests (#323) and added ZIP upload support for skills via the `add_skill` API (#312).
-- Configuration Extensions: Added timeout support in `ovcli.conf` (#308) and fixed agent_id issues in the Rust CLI (#308).
-- Version Support: Added the `--version` flag to `openviking-server` (#358) for easy version validation.
-
-## What's Changed
-* docs : update wechat
-* feat: 多租户 Phase 1 - API 层多租户能力
-* 增加openviking/eval模块,用于评估测试
-* feat(vectordb): integrate KRL for ARM Kunpeng vector search optimization
-* feat: concurrent embedding, GitHub ZIP download, read offset/limit
-* fix: claude code memory-plugin example:add_message改为写入TextPart列表,避免session解析异常
-* Feat/add parts support to http api
-* tests(parsers): add unit tests for office extensions within add_resou…
-* feat: break change, remove is_leaf scalar and use level instead
-* fix(api): complete parts support in SDK layers and simplify error handling
-* feat: support cloud vikingDB
-* Fix image_summary bug
-* feat(config): expose embedding.max_concurrent and vlm.max_concurrent …
-* Multi tenant
-* feat: allow private gitlab domain for code repo
-* fix(storage): idempotent rm/mv operations with vector index sync
-* refactor(filter): replace prefix operator with must
-* fix: VikingDB volcengine URI prefix loss and stale is_leaf filter
-* fix(vectordb): default x86 build to AVX2 and disable AVX512
-* 对齐 docs 和代码里 storage config 的 workspace 属性
-* fix: correct storage.update() call signature in _update_active_counts()
-* feat(memory): add hotness scoring for cold/hot memory lifecycle
-* feat(client): ovcli.conf 支持 timeout 配置 + 修复 Rust CLI agent_id
-* fix(server): 未配置 root_api_key 时仅允许 localhost 绑定
-* fix: resolve Gemini 404, directory collision, and Unicode decoding er…
+## v0.1.18 (2026-02-23)
-## [email protected] (2026-02-27)
+此版本为 OpenViking 带来重大新能力。引入高性能 Rust CLI 和终端文件系统浏览器 UI。文件解析大幅扩展,支持 Word、PowerPoint、Excel、EPub 和 ZIP 格式。新增多 provider 支持用于 embedding 和 VLM 后端。Memory 处理重新设计为具有冲突感知的去重和新抽取流程。
-# OpenViking CLI v0.2.0
+### 重点更新
-### Quick Install (macOS/Linux)
+- **Rust CLI**:全新高速 CLI 实现。
+- **文件解析器**:通过 markitdown 风格解析器支持 Word、PowerPoint、Excel、EPub、ZIP。
+- **TUI**:基础终端 UI 文件系统导航(`ov tui`)。
+- **多 Provider**:支持多个 embedding 和 VLM provider。
+- **Memory**:重新设计的抽取和去重流程,具备冲突感知能力。
+- **Skills**:新增 memory、resource 和 search skills;改进 skill 搜索排序。
+- **目录解析**:新增目录级解析支持。
-### Manual Installation
-Download the appropriate binary for your platform below, extract it, and add it to your PATH.
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.17...v0.1.18)
-The CLI command is simply `ov`:
+## [email protected] (2026-02-14)
-### Checksums
-SHA256 checksums are provided for each binary for verification.
+初始 CLI 二进制发布,跨平台支持 macOS 和 Linux,提供独立的 OpenViking 服务管理和资源操作可执行文件。
-## v0.1.18 (2026-02-23)
+[完整变更记录](https://github.com/volcengine/OpenViking/releases/tag/cli%400.1.0)
-## What's Changed
-* feat: add Rust CLI implementation [very fast]
-* feat: make -o and --json global param
-* feat: provide a test_ov.sh scripts as reference
-* fix: short markdown parse filename
-* Update 03-quickstart-server.md-en version
-* feat: add markitdown-inspired file parsers (Word, PowerPoint, Excel, EPub, ZIP)
-* feat: rename for consistency
-* 上传最新的微信群二维码
-* Update 03-quickstart-server.md
-* Update 01-about-us.md-微信群二维码地址更新
-* fix: build target name
-* fix: fix handle github url right
-* Update README_CN.md-新增云端部署跳转链接与说明
-* Update README.md-新增英文版readme云端部署内容
-* Update 01-about-us.md-英文版微信群二维码更新
-* fix: fix the difference of python CLI and rust CLI
-* Feat/support multi providers , OpenViking支持多providers
-* Fix: 修复 rust CLI 命令中与 python CLI ls/tree 命令不一致的部分;add node limit arg
-* feat: add_memory cli with ov-memory SKILL
-* feat: add directory parsing support to OpenViking
-* fix: auto-rename on duplicate filename conflict
-* fix: guard against None candidate in search_by_id
-* fix bugs 修复provider为openai,但api_key并不是https://api.openai.com/v1而引起的大模型调用失败情况
-* Update open_viking_config.py
-* fix(parser): hash & shorten filenames that exceed filesystem limit
-* test: add comprehensive edge case tests
-* fix: invalid Go version format in agfs-server go.mod
-* fix: add input validation to search_by_id method
-* fix: improve binary content detection and null byte handling
-* suggestion: align grep and glob options
-* suggestion: fix empty in --simple ls response
-* feat: support basic ov tui for fs navigator, boost version 0.2.0
-* build(deps): bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* build(deps): bump actions/checkout from 4 to 6 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 4 to 6 by @dependabot[bot] in
-* refactor(memory): redesign extraction/dedup flow and add conflict-a…
-* fix: target directories retrieve
-* fix: skill search ranking - use overview for embedding and fix visited set filtering
-* fix: use frontmatter description for skill vectorization instead of overview
-* build(deps): bump actions/cache from 4 to 5 by @dependabot[bot] in
-* feat: update media parsers
-* fix: update memory
-* fix: resolve -o option conflict between global output and filesystem output-format
-* feat: add memory, resource and search skills
+## v0.1.17 (2026-02-14)
-## [email protected] (2026-02-14)
+稳定性修复版本。因不稳定性回滚了 VectorDB 中的动态 project name 配置,修复 CI workspace 清理,解决 tree URI 输出错误并增加启动时 `ov.conf` 校验。
-# OpenViking CLI v0.1.0
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.16...v0.1.17)
-### Quick Install (macOS/Linux)
+## v0.1.16 (2026-02-13)
-### Manual Installation
-Download the appropriate binary for your platform below, extract it, and add it to your PATH.
+聚焦 bug 修复与改进的版本。修复 VectorDB 连接问题和 uvloop 与 nest_asyncio 的服务器冲突。临时 URI 现在可读,resource add 超时增大,为 VectorDB 和 Volcengine 后端引入动态 project name 配置。
-The CLI command is simply `ov`:
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.15...v0.1.16)
-### Checksums
-SHA256 checksums are provided for each binary for verification.
+## v0.1.15 (2026-02-13)
-## v0.1.17 (2026-02-14)
+此版本聚焦架构重构和可靠性改进。HTTP 客户端拆分为独立的嵌入和 HTTP 模式以实现更清晰的关注点分离。通过目录重组提升 CLI 启动速度。解决 VectorDB timestamp 和 collection 创建 bug。
-## What's Changed
-* Revert "feat: support dynamic project_name config in VectorDB / volcengine"
-* Fix/ci clean workspace
-* fix: tree uri output error, and validate ov.conf before start
+### 重点更新
-## v0.1.16 (2026-02-13)
+- **重构**:HTTP 客户端拆分为嵌入和 HTTP 模式;QueueManager 从 VikingDBManager 解耦。
+- **CLI**:更快的启动速度;改进 `ls` 和 `tree` 输出。
+- **VectorDB**:修复 timestamp 格式和 collection 创建问题。
+- **解析器**:支持仓库分支和 commit 引用。
+- **OpenClaw**:初步适配 memory 输出语言管线。
-## What's Changed
-* fix: fix vectordb
-* feat: make temp uri readable, and enlarge timeout of add-resource
-* feat: support dynamic project_name config in VectorDB / volcengine
-* fix: server uvloop conflicts with nest_asyncio
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.14...v0.1.15)
-## v0.1.15 (2026-02-13)
+## v0.1.14 (2026-02-12)
-## What's Changed
-
-Now you can try Server/CLI mode!
-
-* refactor(client): 拆分 HTTP 客户端,分离嵌入模式与 HTTP 模式
-* Transaction store
-* fix CI: correct patch targets in test_quick_start_lite.py
-* fix/lifecycle
-* Fix/lifecycle
-* refactor: decouple QueueManager from VikingDBManager
-* refactor: to accelerate cli launch speed, refactor openviking_cli dir
-* efactor: to accelerate cli launch speed, refactor openviking_cli dir
-* fix(vectordb): resolve timestamp format and collection creation issues
-* doc: add multi-tenant-design
-* fix: fix vectordb sparse
-* [WIP]adapt for openclaw: add memory output language pipeline
-* CLI commands (ls, tree) output optimize
-* fix: replace bare except with Exception in llm utils
-* feat(parser): support repo branch and commit refs
-* fix: temp dir check failed
+重大基础设施版本。引入 HTTP Server 和 Python HTTP Client,实现 OpenViking 服务的远程访问。OpenClaw skill 新增 MCP 集成支持。目录预扫描校验、DAG 触发 embedding 和并行资源添加提升了性能和可靠性。
-## v0.1.14 (2026-02-12)
+### 重点更新
-## What's Changed
-* build(deps): bump protobuf from 6.33.2 to 6.33.5 by @dependabot[bot] in
-* refactor: cpp bytes rows
-* fix: agfs port
-* refactor: refactor agfs s3 backend config
-* fix(depends): 修复pip依赖的安全漏洞
-* feat: add HTTP Server and Python HTTP Client (T2 & T4)
-* Add OpenClaw skill for OpenViking MCP integration
-* docs: update docs and github workflows python version, python>=3.10
-* Doc: suggest use ~/.openviking/ov.conf as a default configure path
-* Parallel add
-* feat(directory-scan): add directory pre-scan validation module with f…
-* fix(docs): fix relative paths in README files to match actual docs st…
-* feat: use a default dir ~/.openviking to store configuration, fix win…
-* feat: dag trigger embedding
-* fix: fix windows
-* fix: release py3.13
-* fix: remove await asyncio and call agfs directly
-* fix(depends): 修复测试引入的依赖
-* add_message改为写入TextPart列表,避免session消息解析异常; extract_session增加json转换方法
-* feat: 新增 Bash CLI 基础框架与完整命令实现 (T3 + T5)
-* fix: test case
-* fix: fix release
+- **HTTP Server**:新的服务模式,提供 Python HTTP Client 用于远程访问。
+- **OpenClaw Skill**:OpenViking 的 MCP 集成。
+- **CLI**:完整的 Bash CLI 框架和全面的命令实现。
+- **Embedding**:DAG 触发 embedding 和并行 add 支持。
+- **目录扫描**:新增预扫描校验模块。
+- **配置**:默认配置目录设为 `~/.openviking`。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.12...v0.1.14)
## v0.1.12 (2026-02-09)
-## What's Changed
-* feat: add search_with_sparse_logit_alpha
-* refactor: Refactor S3 configuration structure and fix Python 3.9 compatibility issues
-* fix: fix ci
-* refactor: unify async execution utilities into run_async
-* docs: update community link
-* build(deps): bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* build(deps): bump actions/github-script from 7 to 8 by @dependabot[bot] in
-* build(deps): bump actions/checkout from 4 to 6 by @dependabot[bot] in
-* build(deps): bump actions/setup-go from 5 to 6 by @dependabot[bot] in
-* build(deps): bump actions/upload-artifact from 4 to 6 by @dependabot[bot] in
-* feat: in chatmem example, add /time and /add_resource command
-* Feature/new demo
-* Feat:支持原生部署的vikingdb
-* WIP: fix: run memex locally for #78
-* feat(parse): extract shared upload utilities
-* docs: fiix related document links in /openviking/parse/parsers/README.md
-* fix(parse): prevent Zip Slip path traversal in _extract_zip (CWE-22)
-* Path filter
-* feat: use tabulate for observer
-* fix(parser): fix temporary file leak in HTMLParser download link hand…
-* perf: reuse query embeddings in hierarchical retriever
-* fix(agfs): close socket on error path in _check_port_available
-* fix(storage): make VikingFS.mkdir() actually create the target directory
-* fix: fix sparse
-* feat: support query in mcp, tested with kimi
-* fix: ignore TestVikingDBProject
+此版本改进搜索质量、存储可靠性和代码可维护性。新增 sparse logit alpha 搜索增强检索。在 hierarchical retriever 中复用查询 embedding 提升性能。支持原生 VikingDB 部署。修补了一个严重的 Zip Slip 路径穿越漏洞 (CWE-22)。
+
+### 重点更新
+
+- **搜索**:Sparse logit alpha 支持和优化的查询 embedding 复用。
+- **VikingDB**:原生部署支持。
+- **安全**:Zip Slip 路径穿越修复 (CWE-22)。
+- **重构**:统一异步执行工具;重构 S3 配置。
+- **MCP**:新增查询支持并通过 Kimi 验证。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.11...v0.1.12)
## v0.1.11 (2026-02-05)
-## What's Changed
-* support small github code repos
+新增对小型 GitHub 代码仓库的导入支持,使 OpenViking 能够直接索引和搜索公开代码库。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.10...v0.1.11)
## v0.1.10 (2026-02-05)
-## What's Changed
-* Fix compile
-* fix: fix windows release
+修复编译错误和 Windows 二进制发布打包问题的补丁版本。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/compare/v0.1.9...v0.1.10)
## v0.1.9 (2026-02-05)
-## What's Changed
-* Bump github/codeql-action from 3 to 4 by @dependabot[bot] in
-* Bump actions/setup-go from 5 to 6 by @dependabot[bot] in
-* Bump actions/setup-python from 5 to 6 by @dependabot[bot] in
-* Bump astral-sh/setup-uv from 4 to 7 by @dependabot[bot] in
-* Bump actions/download-artifact from 4 to 7 by @dependabot[bot] in
-* fix session test
-* feat: add GitHub issue and PR templates
-* fix: fix_build
-* docs: update readme
-* fix: Downgraded a log message from warning to info when the Rerank cl…
-* docs: update_readme
-* Remove agfs
-* docs: faq
-* fix: fix agfs-server for windows
-* Upgrade go
-* feat: add more visual example `query`
-* fix: support intel mac install
-* feat: rename backend to provider for embedding and vlm
-* Lint code
-* refactor: optimized pyproject.toml with optional groups
-* feat: linux compile
-* Change 'provider' to 'backend' in configuration
-* update version
-* Revert "Change 'provider' to 'backend' in configuration"
-* chore: use standard logging pkg
-* feat: chat & chat w/ mem examples
-* feat:常规环境下,增加python3.13适配
-* docs: add server cli design
-* refactor: refactor ci action
-* refactor: extract Service layer from async_client
-* fix: fix ci action
-* fix: simplify memory dedup decisions and fix retrieval recursion bug
-* Fix ci action
-* fix: 修复s3fs适配
-* refactor: extract ObserverService from DebugService for cleaner status access API
-* fix: fix release action
+OpenViking 的初始公开发布。此版本建立了核心项目结构,支持 Linux 和 Intel Mac 跨平台。引入服务层架构,将 embedding 和 VLM 后端分离为可配置的 provider。改进了 Memory 去重并修复了检索递归 bug。包含 Python 3.13 兼容性、S3FS 支持以及 chat 和 memory 工作流的使用示例。
+
+### 重点更新
+
+- **初始发布**:核心 OpenViking server、client 和 CLI 基础。
+- **Provider**:可配置的 embedding 和 VLM 后端,provider 抽象层。
+- **架构**:从 async client 中提取 Service 层;ObserverService 从 DebugService 分离。
+- **平台**:Linux 编译支持、Intel Mac 兼容性、Python 3.13 支持。
+- **Memory**:简化去重逻辑并修复检索递归 bug。
+- **示例**:Chat 和 chat-with-memory 使用示例。
+
+[完整变更记录](https://github.com/volcengine/OpenViking/releases/tag/v0.1.9)
diff --git a/docs/zh/about/03-roadmap.md b/docs/zh/about/03-roadmap.md
index a1213d697..966055277 100644
--- a/docs/zh/about/03-roadmap.md
+++ b/docs/zh/about/03-roadmap.md
@@ -9,68 +9,105 @@
- Viking URI 寻址系统
- 双层存储(AGFS + 向量索引)
- 异步/同步客户端支持
+- QueueFS SQLite 存储后端
### 资源管理
- 文本资源管理(Markdown、HTML、PDF)
- 自动 L0/L1 生成
- 带向量索引的语义搜索
- 资源关联和链接
+- 内容写入 API
+- Agent 命名空间管理
+
+### 多模态解析
+- 图像 OCR 和解析
+- 音频转写(Whisper ASR)
+- 视频解析
+- PDF 书签提取
+- Word、PowerPoint、Excel、EPub、ZIP 解析器
+- 代码文件解析
+- 飞书/Lark 文档解析器
### 检索
- 基本语义搜索(`find`)
- 带意图分析的上下文感知搜索(`search`)
- 基于会话的查询扩展
-- 重排序流水线
+- 多供应商重排序流水线(OpenAI、LiteLLM、Cohere、Volcengine)
-### 会话管理
+### 会话与记忆
- 对话状态追踪
- 上下文和技能使用追踪
- 自动记忆提取
- 使用 LLM 的记忆去重
- 会话归档和压缩
+- Working Memory V2 及冷存储归档
### 技能
- 技能定义和存储
- MCP 工具自动转换
- 技能搜索和检索
-### 配置
-- 可插拔的 Embedding 提供者
+### 多租户与安全
+- 多租户支持与账户隔离
+- 文件和文档加密
+- 用户级隐私配置 API
+- API Key 认证
+
+### 配置与供应商
+- 可插拔的 Embedding 提供者(OpenAI、Gemini、Volcengine、MiniMax、LiteLLM、Jina、Cohere、DashScope、Voyage、本地)
- 可插拔的 LLM 提供者
+- 可插拔的重排序提供者
- 基于 YAML 的配置
+- 安装向导(`openviking-server init`)
-### Server & Client 架构
-- HTTP Server (FastAPI)
+### Server 与 Client 架构
+- HTTP Server(FastAPI)
+- 内置 MCP 端点
- Python HTTP Client
-- API Key 认证
- 客户端抽象层(LocalClient / HTTPClient)
+- Web 控制台
+
+### CLI
+- Rust CLI(`ov` 命令)
+- TUI 文件系统浏览器
+- 隐私、搜索、会话、资源及管理命令
+
+### Bot 集成
+- VikingBot 框架
+- 飞书/Lark 频道
+- Telegram 频道
+
+### 生态与插件
+- OpenClaw 插件(编程 Agent 上下文引擎)
+- Claude Code 记忆插件
+- Codex 记忆插件
+
+### 可观测性
+- Prometheus 指标
+- OpenTelemetry 链路追踪
+- HTTP 可观测性中间件
+
+### 部署
+- Docker 镜像和 Docker Compose
+- Kubernetes Helm Chart
+- 云端 VikingDB 支持
---
## 未来计划
-### CLI
-- 完整的命令行界面,支持所有操作
-- 分布式存储后端
-
-### 多模态支持
-- 支持图像、视频、音频等多模态资源的智能解析和访问
-- 多模态资源的目录存储结构
-
### 上下文管理
- 上下文修改对上层的传导更新
-- 支持对上下文的版本管理和回滚 (参考 git)
+- 支持对上下文的版本管理和回滚(参考 git)
-### 资源节点的权限控制
-- 支持多Agent / 多User 使用
-- 不同角色的隔离设计
-- 资源目录节点的访问控制和权限设计
+### 分布式存储
+- 分布式存储后端
-### 生态集成
-- 流行 Agent 框架适配
-- 插件系统,支持自定义组件
+### 生态
+- 更多 Agent 框架适配器
欢迎在 issue 中提出建议和反馈。
+
---
## 贡献