Skip to content

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

fix(session): align async commit API docs and examples (#1188)

Qin Haojiec0b0be4f4c98
diff --git a/.pr_agent.toml b/.pr_agent.toml
index 5396cecba..d01277990 100644
--- a/.pr_agent.toml
+++ b/.pr_agent.toml
@@ -17,8 +17,10 @@
# ---------------------------------------------------------------------------
[config]
output_relevant_configurations = false
-# Use Ark token-billed Chat API instead of Coding Plan.
-model = "doubao-seed-2-0-code-preview-260215"
+# Use Ark token-billed Chat API via LiteLLM's OpenAI-compatible provider.
+# Without the `openai/` prefix, PR-Agent fails with:
+# "LLM Provider NOT provided" and skips review/improve generation.
+model = "openai/doubao-seed-2-0-code-preview-260215"
fallback_models = []
custom_model_max_tokens = 256000
# Use high reasoning for this complex polyglot codebase with subtle async/
diff --git a/docs/en/api/05-sessions.md b/docs/en/api/05-sessions.md
index 485569911..6cc1c470a 100644
--- a/docs/en/api/05-sessions.md
+++ b/docs/en/api/05-sessions.md
@@ -632,7 +632,7 @@ print(f"Task ID: {result['task_id']}")
# Poll background task progress
task = client.get_task(result["task_id"])
if task["status"] == "completed":
- print(f"Memories extracted: {task['result']['memories_extracted']}")
+ print(f"Memories extracted: {sum(task['result']['memories_extracted'].values())}")
```
**HTTP API**
@@ -728,13 +728,20 @@ curl -X GET http://localhost:1933/api/v1/tasks/uuid-xxx \
"result": {
"session_id": "a1b2c3d4",
"archive_uri": "viking://session/a1b2c3d4/history/archive_001",
- "memories_extracted": 5,
+ "memories_extracted": {
+ "profile": 1,
+ "preferences": 2,
+ "entities": 1,
+ "cases": 1
+ },
"active_count_updated": 2
}
}
}
```
+`memories_extracted` in the completed task result reports per-category counts for this commit only. Sum its values when you want the total for this commit.
+
---
## Session Properties
@@ -776,12 +783,14 @@ viking://session/{session_id}/
| Category | Location | Description |
|----------|----------|-------------|
-| profile | `user/memories/.overview.md` | User profile information |
+| profile | `user/memories/profile.md` | User profile information |
| preferences | `user/memories/preferences/` | User preferences by topic |
| entities | `user/memories/entities/` | Important entities (people, projects) |
| events | `user/memories/events/` | Significant events |
| cases | `agent/memories/cases/` | Problem-solution cases |
| patterns | `agent/memories/patterns/` | Interaction patterns |
+| tools | `agent/memories/tools/` | Tool usage knowledge and best practices |
+| skills | `agent/memories/skills/` | Skill execution knowledge and workflow strategies |
---
@@ -828,7 +837,7 @@ print(f"Task ID: {result['task_id']}")
# Optional: poll for completion
task = client.get_task(result["task_id"])
if task and task["status"] == "completed":
- print(f"Memories extracted: {task['result']['memories_extracted']}")
+ print(f"Memories extracted: {sum(task['result']['memories_extracted'].values())}")
client.close()
```
diff --git a/docs/en/concepts/01-architecture.md b/docs/en/concepts/01-architecture.md
index 832208978..f82439f1d 100644
--- a/docs/en/concepts/01-architecture.md
+++ b/docs/en/concepts/01-architecture.md
@@ -58,7 +58,7 @@ OpenViking is a context database designed for AI Agents, unifying all context ty
| **Retrieve** | Context retrieval | Intent analysis (IntentAnalyzer), hierarchical retrieval (HierarchicalRetriever), Rerank |
| **Session** | Session management | Message recording, usage tracking, session compression, memory commit |
| **Parse** | Context extraction | Document parsing (PDF/MD/HTML), tree building (TreeBuilder), async semantic generation |
-| **Compressor** | Memory compression | 6-category memory extraction, LLM deduplication decisions |
+| **Compressor** | Memory compression | 8-category memory extraction, LLM deduplication decisions |
| **Storage** | Storage layer | VikingFS virtual filesystem, vector index, AGFS integration |
## Service Layer
@@ -117,7 +117,7 @@ Messages → Compress → Archive → Memory Extraction → Storage
1. **Messages**: Accumulate conversation messages and usage records
2. **Compress**: Keep recent N rounds, archive older messages
3. **Archive**: Generate L0/L1 for history segments
-4. **Memory Extraction**: Extract 6-category memories from messages
+4. **Memory Extraction**: Extract 8-category memories from messages
5. **Storage**: Write to AGFS + vector index
## Deployment Modes
diff --git a/docs/en/concepts/02-context-types.md b/docs/en/concepts/02-context-types.md
index b5f96f1eb..18454cb5a 100644
--- a/docs/en/concepts/02-context-types.md
+++ b/docs/en/concepts/02-context-types.md
@@ -52,16 +52,18 @@ Memories are divided into user memories and Agent memories, representing learned
- **Dynamic updates**: Continuously updated from interactions by Agent
- **Personalized**: Learned for specific users or specific Agents
-### 6 Categories
+### 8 Categories
| Category | Location | Description | Update Strategy |
|----------|----------|-------------|-----------------|
-| **profile** | `user/memories/.overview.md` | User basic info | ✅ Appendable |
+| **profile** | `user/memories/profile.md` | User basic info | ✅ Merge into one file |
| **preferences** | `user/memories/preferences/` | User preferences by topic | ✅ Appendable |
| **entities** | `user/memories/entities/` | Entity memories (people, projects) | ✅ Appendable |
| **events** | `user/memories/events/` | Event records (decisions, milestones) | ❌ No update |
| **cases** | `agent/memories/cases/` | Learned cases | ❌ No update |
-| **patterns** | `agent/memories/patterns/` | Learned patterns | ❌ No update |
+| **patterns** | `agent/memories/patterns/` | Learned patterns | ✅ Mergeable |
+| **tools** | `agent/memories/tools/` | Tool usage knowledge and best practices | ✅ Mergeable |
+| **skills** | `agent/memories/skills/` | Skill execution knowledge and workflow strategies | ✅ Mergeable |
### Usage
@@ -69,7 +71,8 @@ Memories are divided into user memories and Agent memories, representing learned
# Memories are auto-extracted from sessions
session = client.session()
await session.add_message("user", [{"type": "text", "text": "I prefer dark mode"}])
-await session.commit() # Extracts preference memory
+commit = await session.commit() # Starts background memory extraction
+task = await client.get_task(commit["task_id"]) # Poll until task["status"] == "completed"
# Search memories
results = await client.find(
diff --git a/docs/en/concepts/08-session.md b/docs/en/concepts/08-session.md
index c53bcefcb..0f5fd4f44 100644
--- a/docs/en/concepts/08-session.md
+++ b/docs/en/concepts/08-session.md
@@ -69,6 +69,7 @@ result = session.commit()
# Poll background task progress
task = client.get_task(result["task_id"])
# task["status"]: "pending" | "running" | "completed" | "failed"
+# sum(task["result"]["memories_extracted"].values()): 3
```
## Message Structure
@@ -132,7 +133,7 @@ Unfinished tasks
## Memory Extraction
-### 6 Categories
+### 8 Categories
| Category | Belongs to | Description | Mergeable |
|----------|------------|-------------|-----------|
@@ -142,6 +143,8 @@ Unfinished tasks
| **events** | user | Events/decisions | ❌ |
| **cases** | agent | Problem + solution | ❌ |
| **patterns** | agent | Reusable patterns | ✅ |
+| **tools** | agent | Tool usage knowledge and best practices | ✅ |
+| **skills** | agent | Skill execution knowledge and workflow strategies | ✅ |
### Extraction Flow
@@ -190,7 +193,9 @@ viking://user/memories/
viking://agent/memories/
├── cases/
-└── patterns/
+├── patterns/
+├── tools/
+└── skills/
```
## Related Documents
diff --git a/docs/zh/api/05-sessions.md b/docs/zh/api/05-sessions.md
index 41b5b06a5..c7609a1b3 100644
--- a/docs/zh/api/05-sessions.md
+++ b/docs/zh/api/05-sessions.md
@@ -632,7 +632,7 @@ print(f"Task ID: {result['task_id']}")
# 查询后台任务进度
task = client.get_task(result["task_id"])
if task["status"] == "completed":
- print(f"Memories extracted: {task['result']['memories_extracted']}")
+ print(f"Memories extracted: {sum(task['result']['memories_extracted'].values())}")
```
**HTTP API**
@@ -728,13 +728,20 @@ curl -X GET http://localhost:1933/api/v1/tasks/uuid-xxx \
"result": {
"session_id": "a1b2c3d4",
"archive_uri": "viking://session/a1b2c3d4/history/archive_001",
- "memories_extracted": 5,
+ "memories_extracted": {
+ "profile": 1,
+ "preferences": 2,
+ "entities": 1,
+ "cases": 1
+ },
"active_count_updated": 2
}
}
}
```
+完成态任务结果里的 `memories_extracted` 表示本次 commit 的分类计数;如果只需要本次 commit 的总数,请把这些值求和。
+
---
## 会话属性
@@ -776,12 +783,14 @@ viking://session/{session_id}/
| 分类 | 位置 | 说明 |
|------|------|------|
-| profile | `user/memories/.overview.md` | 用户个人信息 |
+| profile | `user/memories/profile.md` | 用户个人信息 |
| preferences | `user/memories/preferences/` | 按主题分类的用户偏好 |
| entities | `user/memories/entities/` | 重要实体(人物、项目等) |
| events | `user/memories/events/` | 重要事件 |
| cases | `agent/memories/cases/` | 问题-解决方案案例 |
| patterns | `agent/memories/patterns/` | 交互模式 |
+| tools | `agent/memories/tools/` | 工具使用经验与最佳实践 |
+| skills | `agent/memories/skills/` | 技能执行经验与工作流策略 |
---
@@ -828,7 +837,7 @@ print(f"Task ID: {result['task_id']}")
# 可选:等待后台任务完成
task = client.get_task(result["task_id"])
if task and task["status"] == "completed":
- print(f"Memories extracted: {task['result']['memories_extracted']}")
+ print(f"Memories extracted: {sum(task['result']['memories_extracted'].values())}")
client.close()
```
diff --git a/docs/zh/concepts/01-architecture.md b/docs/zh/concepts/01-architecture.md
index a90040e94..ba21f2a00 100644
--- a/docs/zh/concepts/01-architecture.md
+++ b/docs/zh/concepts/01-architecture.md
@@ -57,7 +57,7 @@ OpenViking 是为 AI Agent 设计的上下文数据库,将所有上下文(Me
| **Retrieve** | 上下文检索 | 意图分析(IntentAnalyzer)、层级检索(HierarchicalRetriever)、Rerank 精排 |
| **Session** | 会话管理 | 消息记录、使用追踪、会话压缩、记忆提交 |
| **Parse** | 上下文提取 | 文档解析(PDF/MD/HTML)、树构建(TreeBuilder)、异步语义生成 |
-| **Compressor** | 记忆压缩 | 6 种记忆分类提取、LLM 去重决策 |
+| **Compressor** | 记忆压缩 | 8 种记忆分类提取、LLM 去重决策 |
| **Storage** | 存储层 | VikingFS 虚拟文件系统、向量索引、AGFS 集成 |
## Service 层
@@ -116,7 +116,7 @@ OpenViking 采用双层存储架构,实现内容与索引分离(详见 [存
1. **消息**:累积对话消息和使用记录
2. **压缩**:保留最近 N 轮,旧消息归档
3. **归档**:生成历史片段的 L0/L1
-4. **记忆提取**:从消息中提取 6 种分类记忆
+4. **记忆提取**:从消息中提取 8 种分类记忆
5. **存储**:写入 AGFS + 向量库
## 部署模式
diff --git a/docs/zh/concepts/02-context-types.md b/docs/zh/concepts/02-context-types.md
index c432fad6f..f52ecf6fd 100644
--- a/docs/zh/concepts/02-context-types.md
+++ b/docs/zh/concepts/02-context-types.md
@@ -52,16 +52,18 @@ results = client.find(
- **动态更新:**由 Agent 从交互中持续更新
- **个性化:**针对特定用户或 特定 Agent 学习记录
-### 6 种分类
+### 8 种分类
| 分类 | 位置 | 说明 | 更新策略 |
|------|------|------|----------|
-| **profile** | `user/memories/.overview.md` | 用户基本信息 | ✅ 可追加 |
+| **profile** | `user/memories/profile.md` | 用户基本信息 | ✅ 合并到单文件 |
| **preferences** | `user/memories/preferences/` | 按主题的用户偏好 | ✅ 可追加 |
| **entities** | `user/memories/entities/` | 实体记忆(人物、项目) | ✅ 可追加 |
| **events** | `user/memories/events/` | 事件记录(决策、里程碑) | ❌ 不更新 |
| **cases** | `agent/memories/cases/` | 学习的案例 | ❌ 不更新 |
-| **patterns** | `agent/memories/patterns/` | 学习的模式 | ❌ 不更新 |
+| **patterns** | `agent/memories/patterns/` | 学习的模式 | ✅ 可合并 |
+| **tools** | `agent/memories/tools/` | 工具使用经验与最佳实践 | ✅ 可合并 |
+| **skills** | `agent/memories/skills/` | 技能执行经验与工作流策略 | ✅ 可合并 |
### 使用
@@ -69,7 +71,8 @@ results = client.find(
# 记忆从会话中自动提取
session = client.session()
await session.add_message("user", [{"type": "text", "text": "我喜欢深色模式"}])
-await session.commit() # 提取偏好记忆
+commit = await session.commit() # 启动后台记忆提取
+task = await client.get_task(commit["task_id"]) # 轮询直到 task["status"] == "completed"
# 搜索记忆
results = await client.find(
diff --git a/docs/zh/concepts/08-session.md b/docs/zh/concepts/08-session.md
index e5959059f..94bb11025 100644
--- a/docs/zh/concepts/08-session.md
+++ b/docs/zh/concepts/08-session.md
@@ -69,6 +69,7 @@ result = session.commit()
# 查询后台任务进度
task = client.get_task(result["task_id"])
# task["status"]: "pending" | "running" | "completed" | "failed"
+# sum(task["result"]["memories_extracted"].values()): 3
```
## 消息结构
@@ -132,7 +133,7 @@ commit() 分两阶段执行:
## 记忆提取
-### 6 种分类
+### 8 种分类
| 分类 | 归属 | 说明 | 可合并 |
|------|------|------|--------|
@@ -142,6 +143,8 @@ commit() 分两阶段执行:
| **events** | user | 事件/决策 | ❌ |
| **cases** | agent | 问题+解决方案 | ❌ |
| **patterns** | agent | 可复用流程 | ✅ |
+| **tools** | agent | 工具使用经验与最佳实践 | ✅ |
+| **skills** | agent | 技能执行经验与工作流策略 | ✅ |
### 提取流程
@@ -150,19 +153,20 @@ commit() 分两阶段执行:
↓
向量预过滤 → 找相似记忆
↓
-LLM 去重决策 → CREATE/UPDATE/MERGE/SKIP
+LLM 去重决策 → candidate(skip/create/none) + item(merge/delete)
↓
写入 AGFS → 向量化
```
### 去重决策
-| 决策 | 说明 |
-|------|------|
-| `CREATE` | 新记忆,直接创建 |
-| `UPDATE` | 更新现有记忆 |
-| `MERGE` | 合并多条记忆 |
-| `SKIP` | 完全重复,跳过 |
+| 层级 | 决策 | 说明 |
+|------|------|------|
+| Candidate | `skip` | 候选记忆重复,直接跳过 |
+| Candidate | `create` | 创建候选记忆;必要时先删除冲突旧记忆 |
+| Candidate | `none` | 不创建候选记忆,只处理已有记忆 |
+| Existing item | `merge` | 将候选内容合并到指定已有记忆 |
+| Existing item | `delete` | 删除冲突的已有记忆 |
## 存储结构
@@ -189,7 +193,9 @@ viking://user/memories/
viking://agent/memories/
├── cases/
-└── patterns/
+├── patterns/
+├── tools/
+└── skills/
```
## 相关文档
diff --git a/examples/openclaw-plugin/client.ts b/examples/openclaw-plugin/client.ts
index 56440356c..375a68937 100644
--- a/examples/openclaw-plugin/client.ts
+++ b/examples/openclaw-plugin/client.ts
@@ -456,8 +456,9 @@ export class OpenVikingClient {
if (!task) break;
if (task.status === "completed") {
const taskResult = (task.result ?? {}) as Record<string, unknown>;
+ const memoriesExtracted = (taskResult.memories_extracted ?? {}) as Record<string, number>;
result.status = "completed";
- result.memories_extracted = (taskResult.memories_extracted ?? {}) as Record<string, number>;
+ result.memories_extracted = memoriesExtracted;
return result;
}
if (task.status === "failed") {
diff --git a/examples/openclaw-plugin/context-engine.ts b/examples/openclaw-plugin/context-engine.ts
index 5ebc1d34f..aef6ae25d 100644
--- a/examples/openclaw-plugin/context-engine.ts
+++ b/examples/openclaw-plugin/context-engine.ts
@@ -846,7 +846,7 @@ export function createMemoryOpenVikingContextEngine(params: {
status: commitResult.status,
archived: commitResult.archived ?? false,
taskId: commitResult.task_id ?? null,
- extractedMemories: (commitResult as any).extracted_memories ?? null,
+ extractedMemories: totalExtractedMemories(commitResult.memories_extracted),
});
if (commitResult.task_id && cfg.logFindRequests) {
logger.info(
diff --git a/examples/opencode-memory-plugin/README.md b/examples/opencode-memory-plugin/README.md
index f2d373cd0..538c01aca 100644
--- a/examples/opencode-memory-plugin/README.md
+++ b/examples/opencode-memory-plugin/README.md
@@ -148,7 +148,7 @@ Parameters:
- `session_id?`: optional explicit OpenViking session ID
-Returns background task progress or completion details, including `task_id`, `memories_extracted`, and `archived`.
+Returns background task progress or completion details, including `task_id`, per-category `memories_extracted`, and `archived`.
## Usage Examples
diff --git a/examples/opencode-memory-plugin/README_CN.md b/examples/opencode-memory-plugin/README_CN.md
index 957ebe611..7d7167164 100644
--- a/examples/opencode-memory-plugin/README_CN.md
+++ b/examples/opencode-memory-plugin/README_CN.md
@@ -148,7 +148,7 @@ export OPENVIKING_API_KEY="your-api-key-here"
- `session_id?`:可选的显式 OpenViking 会话 ID
-返回后台任务进度或完成详情,包括 `task_id`、`memories_extracted` 和 `archived`。
+返回后台任务进度或完成详情,包括 `task_id`、分类计数 `memories_extracted` 和 `archived`。
## 使用示例
diff --git a/examples/opencode-memory-plugin/openviking-memory.ts b/examples/opencode-memory-plugin/openviking-memory.ts
index 6f2ef6b8b..829cd1106 100644
--- a/examples/opencode-memory-plugin/openviking-memory.ts
+++ b/examples/opencode-memory-plugin/openviking-memory.ts
@@ -305,12 +305,14 @@ interface SearchResult {
query_plan?: string
}
+type MemoryCounts = number | Record<string, number>
+
interface CommitResult {
session_id: string
status: string
- memories_extracted: number
- active_count_updated: number
- archived: boolean
+ memories_extracted?: MemoryCounts
+ active_count_updated?: number
+ archived?: boolean
task_id?: string
message?: string
stats?: {
@@ -334,7 +336,7 @@ interface TaskResult {
resource_id?: string
result?: {
session_id?: string
- memories_extracted?: number
+ memories_extracted?: MemoryCounts
archived?: boolean
}
error?: string | null
@@ -355,6 +357,27 @@ const DEFAULT_CONFIG: OpenVikingConfig = {
}
}
+function totalMemoriesExtracted(memories?: MemoryCounts): number {
+ if (typeof memories === "number") {
+ return memories
+ }
+ if (!memories || typeof memories !== "object") {
+ return 0
+ }
+ return Object.entries(memories).reduce((sum, [key, value]) => {
+ if (key === "total") {
+ return sum
+ }
+ return sum + (typeof value === "number" ? value : 0)
+ }, 0)
+}
+
+function totalMemoriesFromResult(result?: {
+ memories_extracted?: MemoryCounts
+} | null): number {
+ return totalMemoriesExtracted(result?.memories_extracted)
+}
+
function loadConfig(): OpenVikingConfig {
const configPath = path.join(pluginFileDir, "openviking-config.json")
@@ -788,7 +811,7 @@ async function runSynchronousCommit(
log("INFO", "session", "OpenViking synchronous commit completed", {
openviking_session: mapping.ovSessionId,
opencode_session: opencodeSessionId,
- memories_extracted: result?.memories_extracted ?? 0,
+ memories_extracted: totalMemoriesFromResult(result),
archived: result?.archived ?? false,
})
@@ -981,7 +1004,7 @@ async function pollCommitTaskOnce(
}
if (task.status === "completed") {
- const memoriesExtracted = task.result?.memories_extracted ?? 0
+ const memoriesExtracted = totalMemoriesFromResult(task.result)
const archived = task.result?.archived ?? false
log("INFO", "session", "OpenViking background commit completed", {
@@ -1060,7 +1083,7 @@ async function waitForCommitCompletion(
const task = unwrapResponse(response)
if (task.status === "completed") {
- const memoriesExtracted = task.result?.memories_extracted ?? 0
+ const memoriesExtracted = totalMemoriesFromResult(task.result)
const archived = task.result?.archived ?? false
await finalizeCommitSuccess(mapping, opencodeSessionId, config)
@@ -1694,12 +1717,13 @@ export const OpenVikingMemoryPlugin = async (input: PluginInput): Promise<Hooks>
context.abort,
)
if (task?.status === "completed") {
+ const memoriesExtracted = totalMemoriesFromResult(task.result)
return JSON.stringify(
{
- message: `Memory extraction complete: ${task.result?.memories_extracted ?? 0} memories extracted`,
+ message: `Memory extraction complete: ${memoriesExtracted} memories extracted`,
session_id: task.result?.session_id ?? sessionId,
status: task.status,
- memories_extracted: task.result?.memories_extracted ?? 0,
+ memories_extracted: memoriesExtracted,
archived: task.result?.archived ?? false,
task_id: task.task_id,
},
@@ -1729,12 +1753,13 @@ export const OpenVikingMemoryPlugin = async (input: PluginInput): Promise<Hooks>
}
if (commitStart.mode === "completed") {
+ const memoriesExtracted = totalMemoriesFromResult(commitStart.result)
return JSON.stringify(
{
- message: `Memory extraction complete: ${commitStart.result.memories_extracted ?? 0} memories extracted`,
+ message: `Memory extraction complete: ${memoriesExtracted} memories extracted`,
session_id: commitStart.result.session_id ?? sessionId,
status: commitStart.result.status ?? "completed",
- memories_extracted: commitStart.result.memories_extracted ?? 0,
+ memories_extracted: memoriesExtracted,
archived: commitStart.result.archived ?? false,
},
null,
@@ -1762,12 +1787,13 @@ export const OpenVikingMemoryPlugin = async (input: PluginInput): Promise<Hooks>
)
}
+ const memoriesExtracted = totalMemoriesFromResult(task.result)
return JSON.stringify(
{
- message: `Memory extraction complete: ${task.result?.memories_extracted ?? 0} memories extracted`,
+ message: `Memory extraction complete: ${memoriesExtracted} memories extracted`,
session_id: task.result?.session_id ?? sessionId,
status: task.status,
- memories_extracted: task.result?.memories_extracted ?? 0,
+ memories_extracted: memoriesExtracted,
archived: task.result?.archived ?? false,
task_id: task.task_id,
},
diff --git a/openviking/session/compressor.py b/openviking/session/compressor.py
index e2e90be17..46b56ed41 100644
--- a/openviking/session/compressor.py
+++ b/openviking/session/compressor.py
@@ -4,7 +4,7 @@
Session Compressor for OpenViking.
Handles extraction of long-term memories from session conversations.
-Uses MemoryExtractor for 6-category extraction and MemoryDeduplicator for LLM-based dedup.
+Uses MemoryExtractor for 8-category extraction and MemoryDeduplicator for LLM-based dedup.
"""
from dataclasses import dataclass
@@ -57,7 +57,7 @@ class ExtractionStats:
class SessionCompressor:
- """Session memory extractor with 6-category memory extraction."""
+ """Session memory extractor with 8-category memory extraction."""
def __init__(
self,
diff --git a/openviking/session/memory_extractor.py b/openviking/session/memory_extractor.py
index 820111638..3c84813ed 100644
--- a/openviking/session/memory_extractor.py
+++ b/openviking/session/memory_extractor.py
@@ -3,9 +3,9 @@
"""
Memory Extractor for OpenViking.
-Extracts 6 categories of memories from session:
+Extracts 8 categories of memories from session:
- UserMemory: profile, preferences, entities, events
-- AgentMemory: cases, patterns
+- AgentMemory: cases, patterns, tools, skills
"""
import re
@@ -99,7 +99,7 @@ class MergedMemoryPayload:
class MemoryExtractor:
- """Extracts memories from session messages with 6-category classification."""
+ """Extracts memories from session messages with 8-category classification."""
# Category to directory mapping
CATEGORY_DIRS = {
diff --git a/tests/integration/test_compressor_v2_e2e.py b/tests/integration/test_compressor_v2_e2e.py
index 983024cbb..e5a71598f 100644
--- a/tests/integration/test_compressor_v2_e2e.py
+++ b/tests/integration/test_compressor_v2_e2e.py
@@ -8,6 +8,7 @@ Uses AsyncHTTPClient to connect to local openviking-server at 127.0.0.1:1933.
No need to worry about ov.conf - server uses its own config.
"""
+import asyncio
from dataclasses import asdict
from datetime import datetime
@@ -24,6 +25,16 @@ logger = get_logger(__name__)
SERVER_URL = "http://127.0.0.1:1933"
+async def _wait_for_task(client: AsyncHTTPClient, task_id: str, timeout: float = 60.0) -> dict:
+ """Wait for a background session commit task to finish."""
+ for _ in range(int(timeout / 0.1)):
+ task = await client.get_task(task_id)
+ if task and task["status"] in {"completed", "failed"}:
+ return task
+ await asyncio.sleep(0.1)
+ raise TimeoutError(f"Task {task_id} did not complete within {timeout}s")
+
+
def create_test_conversation_messages():
"""Create a conversation that should trigger memory extraction (Chinese with various memory types)"""
return [
@@ -151,8 +162,11 @@ class TestCompressorV2EndToEnd:
# 3. Commit session (this should trigger memory extraction)
print("\nCommitting session...")
commit_result = await client.commit_session(session_id)
- assert commit_result["status"] == "committed"
+ assert commit_result["status"] == "accepted"
+ assert commit_result["task_id"] is not None
print(f"Commit result: {commit_result}")
+ task_result = await _wait_for_task(client, commit_result["task_id"])
+ assert task_result["status"] == "completed"
# 4. Wait for memory extraction to complete
print("\nWaiting for processing...")
diff --git a/tests/integration/test_full_workflow.py b/tests/integration/test_full_workflow.py
index 569b70617..5906af8bb 100644
--- a/tests/integration/test_full_workflow.py
+++ b/tests/integration/test_full_workflow.py
@@ -113,7 +113,8 @@ class TestSessionWorkflow:
# 5. Commit
commit_result = session.commit()
- assert commit_result["status"] == "committed"
+ assert commit_result["status"] == "accepted"
+ assert commit_result["task_id"] is not None
# 6. Wait for memory extraction
await client.wait_processed()
@@ -127,7 +128,9 @@ class TestSessionWorkflow:
session1 = client.session(session_id=session_id)
session1.add_message("user", [TextPart("First message")])
session1.add_message("assistant", [TextPart("First response")])
- session1.commit()
+ commit_result1 = session1.commit()
+ assert commit_result1["status"] == "accepted"
+ assert commit_result1["task_id"] is not None
# 2. Reload session
session2 = client.session(session_id=session_id)
@@ -139,7 +142,8 @@ class TestSessionWorkflow:
# 4. Commit again
commit_result = session2.commit()
- assert commit_result["status"] == "committed"
+ assert commit_result["status"] == "accepted"
+ assert commit_result["task_id"] is not None
class TestImportExportWorkflow:
@@ -228,7 +232,8 @@ class TestFullEndToEndWorkflow:
# Commit session
commit_result = session.commit()
- assert commit_result["status"] == "committed"
+ assert commit_result["status"] == "accepted"
+ assert commit_result["task_id"] is not None
# ===== Phase 4: Import/Export =====
if resource_uris:
diff --git a/tests/session/test_session_commit.py b/tests/session/test_session_commit.py
index d70391dd7..1951242e2 100644
--- a/tests/session/test_session_commit.py
+++ b/tests/session/test_session_commit.py
@@ -37,6 +37,7 @@ class TestCommit:
assert result.get("status") == "accepted"
assert "session_id" in result
assert result.get("task_id") is not None
+ assert "memories_extracted" not in result
async def test_commit_extracts_memories(
self, session_with_messages: Session, client: AsyncOpenViking
@@ -49,6 +50,8 @@ class TestCommit:
task_result = await _wait_for_task(task_id)
assert task_result["status"] == "completed"
assert "memories_extracted" in task_result["result"]
+ memory_counts = task_result["result"]["memories_extracted"]
+ assert isinstance(memory_counts, dict)
# Wait for semantic/embedding queues
await client.wait_processed(timeout=60.0)
diff --git a/tests/test_session_task_tracking.py b/tests/test_session_task_tracking.py
index 455db644c..a3c09ea14 100644
--- a/tests/test_session_task_tracking.py
+++ b/tests/test_session_task_tracking.py
@@ -86,7 +86,7 @@ def _make_tracked_commit(behavior="instant", result_overrides=None, gate=None, s
final_result = {
"session_id": _sid,
"archive_uri": archive_uri,
- "memories_extracted": 0,
+ "memories_extracted": {},
"active_count_updated": 0,
}
if result_overrides:
@@ -141,7 +141,7 @@ async def test_task_lifecycle_success(api_client):
service.sessions.commit_async = _make_tracked_commit(
behavior="gated",
- result_overrides={"memories_extracted": 5},
+ result_overrides={"memories_extracted": {"profile": 3, "preferences": 2}},
gate=commit_gate,
started=commit_started,
)
@@ -167,7 +167,7 @@ async def test_task_lifecycle_success(api_client):
assert task_resp.status_code == 200
result = task_resp.json()["result"]
assert result["status"] == "completed"
- assert result["result"]["memories_extracted"] == 5
+ assert result["result"]["memories_extracted"] == {"profile": 3, "preferences": 2}
# ── Task lifecycle: pending → running → failed ──