# mattzcarey/shippie

extendable code review and QA agent 🚢

Repository: https://github.com/mattzcarey/shippie
Canonical: https://ross.abutalabs.com/products/shippie
Homepage: https://shippie.dev
Language: TypeScript
License: MIT
License Family: permissive
Topics: openai, huggingface, opensource, code-quality, code-review, github, agent, agents, mcp, modelcontextprotocol, qa, qa-automation, flue, pi
Last push: 2026-08-12T20:11:18+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 97, release rhythm 90, longevity 82
- inputs: {"age_days": 1154, "days_push": 21, "days_rel": 71, "gap_med": 1.0, "n_releases_24m": 29}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2483, forks 243 (observed 2026-08-28T04:06:55.450935+00:00)

## What it is
Shippie is an extendable AI code-review agent that reads pull request diffs, explores the codebase with developer tools, and posts focused review comments via GitHub Actions, GitLab CI, or locally. It is provider-agnostic (Anthropic, OpenAI, OpenRouter, Cloudflare Workers AI) and can act as an MCP client to reach external tools.

## Use cases
- automatically review pull requests with ai
- detect exposed secrets and bugs in code diffs
- add an ai code reviewer to github actions
- run code review locally on staged changes
- trigger ai review by commenting on a pr
- extend code review agent with mcp tools

## When to choose
- you want automated AI feedback on every pull request in CI
- you need a provider-agnostic reviewer supporting Anthropic, OpenAI, and others
- you want an agent that can use MCP tools like browser automation or observability
- you prefer running reviews locally on staged changes before pushing

## When to avoid
- you need deterministic linting or static analysis rather than LLM-based review
- you cannot share code with external AI providers due to privacy constraints
- you want a fully free solution without LLM API costs
- you need deep IDE-integrated review rather than CI-based comments

## Facets
- artifact type: cli-tool
- maturity: active
- function: code-review, agent-framework, mcp, ci-cd, llm-inference
- domain: developer-tools, version-control, artificial-intelligence
- platform: cli, cloud, self-hosted
- tags: ai-code-review, github-actions, pull-request, qa-automation, mcp-client, typescript, automation, nodejs, docker

## Member repositories
- mattzcarey/shippie (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:55.450935+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 2026-08-30T02:28:06.478315+00:00, confidence not recorded.
  - readme: https://github.com/mattzcarey/shippie (fetched 2026-08-28T04:06:55.450935+00:00, sha 5425fc336b80)
  - homepage: https://shippie.dev (fetched 2026-08-29T10:10:06.494376+00:00, sha 04377d81d47d)
  - registry_npm: https://registry.npmjs.org/shippie (fetched 2026-08-29T10:10:06.503232+00:00, sha f7ce34921751)
- Data as of 2026-08-30T08:39:29.467469+00:00.