🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
TypeScript agentic-aiagentic-frameworkagentic-workflowagentsai-agentsai-assistantai-skillsautonomous-agentsclaudecodexdsh-pluginharnessmcp-servermulti-agentmulti-agent-systemsnpmskillsswarmswarm-intelligencetypescript
release: 3.65.0 (#4595)
rUvd000e27e82c7
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex 07228100bf..bbcc723388 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -5,6 +5,59 @@ All notable changes to the Ruflo project (formerly Claude Flow) are documented hThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).+## [3.65.0] - 2026-10-30++Minor release: `@claude-flow/cli`, `claude-flow`, `ruflo` 3.65.0, plus ruflo-mods plugin 0.12.0. This release ships the eight follow-ups to #4581 on how ruflo integrates with Claude Code mods (ADR-503 to ADR-510, #4585).++### Added+- **Learning-loop gaming defences** (ADR-503, #4590).+ - A verified-success turn is tainted when it edited tests (`edited-tests`), edited a verifier's config (`edited-verifier-config`), or ran a narrowed or weakened check (`narrowed-verifier`).+ - Tainted records never train unless `.claude-flow/config.json` `learningLoop.allowTaint` lists their classes.+ - An opt-in rotating re-verify step (`learningLoop.reverify`, off by default) re-runs an allowlisted check in a temporary worktree.+ - An anomaly gate holds back lessons whose evidence looks gamed.+ - `scripts/eval-learning-loop.mjs` is a manual, paid evaluation harness.+- **Real-binary test tier** (ADR-504, #4589). This runs the real `claude` binary with ruflo-mods against a scripted mock of the Anthropic Messages API, with no API key needed. The contracts check that:+ - every mod loads;+ - the learning loop records outcomes;+ - lessons reach the model;+ - the guard denies risky commands;+ - route context is added.++ It runs as the CI job `mods-real-binary` and as `npm run test:real-binary`.+- **`compactSnapshot`** (ruflo-mods option, default on; ADR-505, #4591).+ - Before compaction, the mod saves a screened transcript to rotating slots under `.claude-flow/sessions/compactions/`.+ - After compaction, it adds one bounded `[RUFLO_AFTER_COMPACT]` block with recent lessons and outcomes, open swarm and claim ids, and the snapshot path.+ - `ruflo doctor --component compact-snapshots` warns when snapshots exist but `.claude-flow/` is not gitignored.+- **Plugin secrets via `userConfig` `sensitive: true`** (ADR-506, #4587).+ - The cost-tracker, metaharness and x-gateway plugins read their keys through a shared `readSecret` helper. It checks the masked plugin option first and falls back to the environment.+ - A new `scripts/audit-plugin-secrets.mjs` CI gate fails on secret-named options that are not marked sensitive, and on literal secrets in manifests.+- **Plugin footprint cut** (ADR-507, #4588).+ - Operational skills are user-invoked only (`disable-model-invocation: true`).+ - Long skill and command descriptions are trimmed to one sentence, with a ceiling enforced by an audit.+ - `scripts/measure-plugin-footprint.mjs` reports the result, with before and after numbers in `v3/docs/validation/`.+- **Cost budgets that can feed a gateway** (ADR-508, #4592).+ - `.claude-flow/cost/budget.json` holds the budget windows.+ - `ruflo cost budget set` writes it.+ - `ruflo cost budget export --gateway litellm [--apply]` produces LiteLLM `budget_limits` and per-model budget fallbacks. `--apply` requires HTTPS except on loopback, and reads the key from an environment variable only.+ - `ruflo cost report` breaks spend down by agent and by tool.+ - New ruflo-mods option `costLedger` (default on).+ - `--gateway meta-llm` refuses until meta-llm documents a budget API.+- **Plugin trust** (ADR-509, #4593).+ - Plugins carry Ed25519 `signature.json` manifests, signed with the existing helpers key.+ - `ruflo plugins verify` checks a plugin against its manifest and reports verified, modified, unsigned or bad-signature.+ - A static scan runs before `mods install`, `init` or `plugins install`. A high finding stops the install unless you pass `--accept-risk <plugin>`.+ - `ruflo init` stops on risky project auto-approve keys (`enableAllProjectMcpServers`, `bypassPermissions`, wildcard Bash and others). `--keep-risky-settings` and `--drop-risky-settings` let you continue.+ - Under `modTrust: refuse-risky`, a modified ruflo plugin is refused.+- **Headless plugin staging** (ADR-510, #4586).+ - `resolvePluginPaths` finds ruflo plugins in a local checkout, the marketplace clone or the install cache.+ - Headless `claude -p` workers now get `--plugin-dir` for ruflo-mods automatically.+ - `ruflo plugins paths [--json]` prints the resolved folders for Agent SDK `{ type: "local", path }` callers.+ - `ruflo doctor --component headless-plugins` checks this setup.++### Not yet+- Per-skill trimming (`skillTrim`) was not built: Claude Code exposes no hook that hides a skill from the model's list.+- The 30-task learning-loop evaluation from #4581 §7 has still not been run. The loop stays on by default, as the maintainers decided in ADR-502 §6.+## [3.64.0] - 2026-10-29Minor release: `@claude-flow/cli`, `claude-flow`, `ruflo` 3.64.0; ruflo-mods plugin 0.8.0.diff --git a/package-lock.json b/package-lock.jsonindex d1912c77eb..25ba0ff443 100644--- a/package-lock.json+++ b/package-lock.json@@ -1,12 +1,12 @@{"name": "claude-flow",- "version": "3.64.0",+ "version": "3.65.0","lockfileVersion": 3,"requires": true,"packages": {"": {"name": "claude-flow",- "version": "3.64.0",+ "version": "3.65.0","bundleDependencies": ["@claude-flow/codex","@claude-flow/mcp",diff --git a/package.json b/package.jsonindex 94c5ed075c..b0234a4e2b 100644--- a/package.json+++ b/package.json@@ -1,6 +1,6 @@{"name": "claude-flow",- "version": "3.64.0",+ "version": "3.65.0","workspaces": ["v3/@claude-flow/codex","v3/@claude-flow/mcp",diff --git a/ruflo/package.json b/ruflo/package.jsonindex bd81ad8b6f..9abdf31ea4 100644--- a/ruflo/package.json+++ b/ruflo/package.json@@ -1,6 +1,6 @@{"name": "ruflo",- "version": "3.64.0",+ "version": "3.65.0","description": "Ruflo - Enterprise AI agent orchestration platform. Deploy 60+ specialized agents in coordinated swarms with self-learning, fault-tolerant consensus, vector memory, and MCP integration","main": "bin/ruflo.js","type": "module",@@ -40,7 +40,7 @@"package:rvf": "bash src/scripts/package-rvf.sh"},"dependencies": {- "@claude-flow/cli": "3.64.0"+ "@claude-flow/cli": "3.65.0"},"optionalDependencies": {},"overrides": {diff --git a/v3/@claude-flow/cli/package.json b/v3/@claude-flow/cli/package.jsonindex c69ce48dcd..04ef25ae6c 100644--- a/v3/@claude-flow/cli/package.json+++ b/v3/@claude-flow/cli/package.json@@ -1,6 +1,6 @@{"name": "@claude-flow/cli",- "version": "3.64.0",+ "version": "3.65.0","type": "module","description": "Ruflo CLI - Enterprise AI agent orchestration with 60+ specialized agents, swarm coordination, MCP server, self-learning hooks, and vector memory for Claude Code","main": "dist/src/index.js",