# mozilla-ai/cq

An open standard for shared agent learning. Agents persist, share, and query collective knowledge so they stop rediscovering the same failures independently.

Repository: https://github.com/mozilla-ai/cq
Canonical: https://ross.abutalabs.com/products/cq
Homepage: https://docs.mozilla.ai/cq
Language: Go
License: Apache-2.0
License Family: permissive
Topics: agents, python, go
Last push: 2026-08-14T18:51:25+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 97, release rhythm 86, longevity 13
- inputs: {"age_days": 182, "days_push": 19, "days_rel": 19, "gap_med": 0, "n_releases_24m": 116}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1258, forks 66 (observed 2026-08-28T04:04:09.514443+00:00)

## What it is
cq is an open standard and tooling for shared agent learning: AI coding agents persist, share, and query structured knowledge units so they avoid repeating each other's mistakes. It ships as a CLI that installs an MCP server and skill into hosts like Claude, Codex, Copilot, Cursor, and OpenCode, with Go and Python SDKs.

## Use cases
- stop my coding agent from rediscovering the same bugs
- share debugging knowledge between AI agents
- query known pitfalls before an agent retries a task
- propose learnings from an agent session automatically
- install an MCP knowledge server into Claude or Cursor
- build a collective knowledge base of undocumented API behavior

## When to choose
- you use AI coding agents and want them to learn from past failures
- you want a shared, queryable knowledge store across multiple agent hosts
- you need an MCP server plus CLI and SDKs for agent knowledge sharing

## When to avoid
- you need a stable, production-hardened system (0.x with breaking changes)
- you want fully automatic knowledge capture without human approval
- you don't use MCP-capable or supported coding agents

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: mcp, search-engine, cli, developer-tools, sdk
- domain: large-language-models, developer-tools
- platform: cli, go, python, cross-platform, windows
- tags: shared-agent-learning, knowledge-store, coding-agents, mcp-server, open-standard, agent-memory, ai-agents, knowledge-management, macos, linux

## Member repositories
- mozilla-ai/cq (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:09.514443+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:05:36.311068+00:00, confidence not recorded.
  - readme: https://github.com/mozilla-ai/cq (fetched 2026-08-28T04:04:09.514443+00:00, sha d8ad1bdcc0c7)
  - homepage: https://docs.mozilla.ai/cq (fetched 2026-08-29T12:17:14.595286+00:00, sha ec0e9c337f7e)
  - site_page: https://docs.mozilla.ai/cq/guides/install (fetched 2026-08-29T12:17:14.604492+00:00, sha 44e666cceb39)
  - site_page: https://docs.mozilla.ai/cq/guides/quickstart (fetched 2026-08-29T12:17:14.606480+00:00, sha 314ae5dd2aeb)
- Data as of 2026-08-30T08:39:29.467469+00:00.