# gossipcat-ai

> Multi-agent code review mesh — orchestrates AI agents from multiple providers to review code in parallel, cross-review each other's findings, and build accuracy profiles over time. Agents that catch r

- **URL**: https://www.freshcrate.ai/projects/gossipcat-ai
- **Author**: gossipcat-ai
- **Category**: MCP Servers
- **Latest version**: `v0.5.3` (2026-06-04)
- **License**: MIT
- **Source**: https://github.com/gossipcat-ai/gossipcat-ai
- **Language**: TypeScript
- **GitHub**: 22 stars, 4 forks
- **Registry**: github (`gossipcat-ai/gossipcat-ai`)
- **Tags**: `agentic`, `claude-code`, `mcp`, `memory`, `openclaw`, `orchestration`, `relay`, `skills`, `typescript`

## Description

Multi-agent code review mesh — orchestrates AI agents from multiple providers to review code in parallel, cross-review each other's findings, and build accuracy profiles over time. Agents that catch real bugs get picked more often. Agents that hallucinate get deprioritized. MCP server for Claude Code, Cursor, and other IDEs.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v0.5.3` | 2026-06-04 | High | ## What's Changed * fix(dashboard-v2): DESIGN.md token sweep — SystemPulse + GraphRail + SkillsApiResponse hoist by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/489 * fix(dashboard-v2): RecentSignalsPeek designer pass — CONF column + skeleton + tokens by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/490 * fix(dashboard-v2+relay): LOW bundle — Sparkline threshold, AGENT_ID_RE dots, stable ribbon keys, drop shadow by @ataberk-xyz in https://github.com/gossi |
| `v0.5.2` | 2026-05-24 | High | ## What's Changed * fix(dashboard-v2 + relay): DESIGN.md cleanup + task-graph rotation reader by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/487 * chore(release): v0.5.2 by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/488   **Full Changelog**: https://github.com/gossipcat-ai/gossipcat-ai/compare/v0.5.1...v0.5.2 |
| `v0.4.28` | 2026-05-19 | High | ## What's Changed * polish(skill-engine): EXCLUDED_EXTS dotfiles + HANDBOOK rationale — follow-ups to #414 by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/416 * fix(citations): preserve Windows drive letter in citation regex — closes #413 by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/417 * fix(consensus-engine): anchor warning false-positive on nested worktrees — closes #401 by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/418 * fix |
| `v0.4.25` | 2026-05-14 | High | ## What's Changed * chore(orchestrator): doc-truth fixes — wilson-score header + cross-review anchor resolutionRoots by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/365 * fix(orchestrator): PR #364/#365 follow-up cluster by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/366 * fix(orchestrator): signal log readers walk rotated .jsonl.1 — recover historical evidence (Phase A) by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/367 * fix(dash |
| `v0.4.24` | 2026-05-08 | High | BM25 memory-search wiring + hono CVE override. Closes the follow-up deferred from v0.4.23 (PR #360 shipped the sidecar; this release wires it into the query path).  ## Added  - **`MemorySearcher.searchCorpus()`** (PR #364). Routes `agent_id="_project"` through the BM25 sidecar built in #360: `tokenize` → `loadIndex` (lazy mtime-keyed incremental rebuild) → `rankDocuments` → `SearchResult` projection. Per-agent knowledge-file path unchanged for non-`_project` queries. 3 new tests cover ranking, e |
| `v0.4.22` | 2026-04-28 | High | **Skill-graduation runtime fix.** Bundles PRs #305, #306, #307 — closes the skill-effectiveness loop end-to-end.  ## What's fixed  - **Effectiveness runner now actually fires after consensus** (#307). Was: synchronous await at handler tail, killed by relay SIGTERM (12-min disconnect cycle) on 9/9 rounds in diagnosis. Now: detached via `setImmediate`, survives MCP response close. - **Path-traversal hardening** on the runner (#307). `agentId` + skill-file `category` now validated against `SAFE_NAM |
| `v0.4.19` | 2026-04-22 | High | Hardening release for the Stage 2 premise-verification pipeline. A 2-agent consensus audit on the merged PRs #241/#242/#243 surfaced three HIGH-severity findings; this release ships the fixes plus one skill-prompt correction and one skill-parser noise fix. All changes ride on top of the existing `modality` signal wiring, which audit confirmed correct.  ### Fixed  - **Claim-verifier path containment + input length caps** (#244). `runRg` and `verifyFileLine` called `resolve(projectRoot, agentInput |
| `v0.4.15` | 2026-04-20 | High | ## What's Changed * test(install): regression guards for tarball integrity + rules drift by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/186 * docs(claude-md): add '## Your Role' heading + restore drift anchor by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/187 * feat(orchestrator,cli): L2 — signal-writer visibility (Symbol accessor + typed helpers + wiring) by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/190 * test(signal-pipeline): |
| `v0.4.14` | 2026-04-19 | High | ## What's Changed * feat(orchestrator): Layer 1 signal-pipeline parity test by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/173 * feat(orchestrator): Layer 3 signal-pipeline drift detector by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/174 * feat(security): gossip_remember + memory_query hardening by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/175 * fix(completion-signals): _system guard + parity length assert + redundant nullish b |
| `v0.4.13` | 2026-04-18 | High | ## What's Changed * fix(signal-pipeline): shared emitCompletionSignals helper + error-path + memoryQueryCalled threading by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/169 * chore(release): v0.4.13 by @ataberk-xyz in https://github.com/gossipcat-ai/gossipcat-ai/pull/170   **Full Changelog**: https://github.com/gossipcat-ai/gossipcat-ai/compare/v0.4.12...v0.4.13 |

## Citation

- HTML: https://www.freshcrate.ai/projects/gossipcat-ai
- Markdown: https://www.freshcrate.ai/projects/gossipcat-ai.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/gossipcat-ai/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._