AgentLint CI

Run AgentLint.io AI agent readiness audits in any CI/CD pipeline.

Platform Guides

| Platform | Integration | Docs | |---|---|---| | GitHub Actions | Native action with PR comments | docs/github.md | | GitLab CI | CLI via npx | docs/gitlab.md | | CircleCI | CLI via npx | docs/circleci.md | | Bitbucket Pipelines | CLI via npx | docs/bitbucket.md |

GitHub Actions

- uses: 1devstudio/agentlint-ci@v1
  with:
    url: "https://your-site.com"
    api-key: ${{ secrets.AGENTLINT_API_KEY }}
    threshold: 50
  env:
    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

CLI (GitLab, CircleCI, Bitbucket, anything else)

npx @1devstudio/agentlint-ci audit --url https://your-site.com --api-key $AGENTLINT_API_KEY --threshold 50

CLI Options

--url <url>          URL to audit (required)
--api-key <key>      API key (or set AGENTLINT_API_KEY env var)
--threshold <n>      Minimum score to pass (default: 0)
--api-url <url>      API base URL (default: https://agentlint.io)
--format <fmt>       Output: text, json, markdown (default: text)

Exit code 1 if score < threshold or on error.

Output Formats

Text (default) — human-readable summary:

AgentLint Audit — 72/100
URL: https://example.com
Framework: Next.js · Render: SSR · Size: 42.1KB · Fetch: 340ms

  ✔ Crawler Access: 90/100
  ⚠ Markdown & LLMs: 60/100
  ✔ Structured Data: 85/100
  ✔ Rendering: 70/100
  ✘ Token Efficiency: 55/100

Report: https://agentlint.io/report/abc123

JSON (--format json) — machine-readable for scripting:

{
  "score": 72,
  "passed": true,
  "threshold": 50,
  "reportUrl": "https://agentlint.io/report/abc123",
  "categories": { "crawlerAccess": 90, "markdownLlms": 60, ... }
}

Markdown (--format markdown) — for posting to PRs/issues.

Get an API Key

1. Sign up at agentlint.io
2. Go to Dashboard → API Keys
3. Create a key and add it as a CI secret

License

MIT