Skip to content

claude-flow

v3.30.2 Breaking

This release includes breaking changes for platform teams planning a safe upgrade.

✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agentic-ai agentic-framework agentic-workflow agents ai-agents ai-assistant
+14 more
ai-coding ai-skills autonomous-agents claude-code codex harness mcp-server multi-agent multi-agent-systems npm skills swarm swarm-intelligence typescript

ReleasePort's take

Light signal
editorial:auto 12d

Version v3.30.2 adds three functional checks to the "ruflo doctor --component memory" command and fixes CI env‑audit allowlist false positives for four environment variables.

Why it matters: The added health checks improve reliability of the memory component, while correcting CI pipeline false‑positive alerts that could block merges; both changes affect developers and SREs monitoring pipelines.

Summary

AI summary

Updates Related, Deferred, and https://github.com/ruvnet/ruflo/issues/2677 across a mixed release.

Changes in this release

Bugfix Medium

Adds three functional checks to "ruflo doctor --component memory".

Adds three functional checks to "ruflo doctor --component memory".

Source: llm_adapter@2026-07-15

Confidence: high

Bugfix Low

Fixes CI env-audit allowlist false positives for four environment variables.

Fixes CI env-audit allowlist false positives for four environment variables.

Source: llm_adapter@2026-07-15

Confidence: high

Full changelog

What

Ships @stuinfla's proposed doctor memory checks 1–3 from #2677. Bug fix release — no new user-facing surface, no breaking changes.

Doctor memory functional checks

Before v3.30.2: `ruflo doctor --component memory` was `existsSync` + `statSync` — could not fail on any file that exists. Reported PASS on a 99.97%-empty and even a SQLite-malformed DB per @stuinfla's 81-store fleet report.

After: existence check still runs (unchanged), plus three new checks that layer functional assertions:

  • Memory Integrity — opens via sql.js + `PRAGMA integrity_check`
  • Memory Content — >=95% of `memory_entries` rows have non-empty content
  • Memory Embedding Coverage — >=95% of populated rows have a vector

Each check prints the exact measurement in the message ("content 3/11133 (0.03%)"), fails with a fix-suggestion, and handles the encrypted-DB case as WARN not PASS (per @stuinfla's "UNKNOWN is never PASS" rule).

Also fixed

CI env-audit allowlist entries for four escape hatches that were flagging on every PR:

  • `RUFLO_AI_BUDGET_DIR` × 3 sites (#2663 repo-supervisor)
  • `RUFLO_AI_BUDGET_DISABLE` (#2663 AI-cost fuse)
  • `RUFLO_METAHARNESS_SKIP_LOCAL` (metaharness invoke shim)
  • `RUFLO_HELPERS_LOCKED` (v3.30.0 .LOCKED marker escape hatch)

Deferred

  • Check 4 (recall probe) — needs write+search+delete round-trip through the CLI's memory pipeline; its own PR
  • Checks 5–8 (distillation, reflexion, skills, continuity) + the underlying distill→reflexion write-path fix — @stuinfla's own PR proposal

Related

  • PR #2681
  • Issue #2677 (still open — checks 4-8 pending)

Install

```bash
npx ruflo@latest doctor --component memory

or

npm install -g @claude-flow/[email protected]
```

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track claude-flow

Get notified when new releases ship.

Sign up free

About claude-flow

Deploy multi-agent swarms with coordinated workflows.

All releases →

Related context

Beta — feedback welcome: [email protected]