nyaomaru/Changelog Bot Runner Nyaomaru
Auto-update CHANGELOG.md from git history and release notes, optionally using an LLM. Creates a PR with the changes.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jul 11, 2026
- License
- MIT
Pinned Snippet
uses: nyaomaru/changelog-bot@6af81a9665def4c947cf54854af4dac0a47dd57f # v0.6.5tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| changelog-path | Path to CHANGELOG file | no | — |
| config-path | Path to changelog-bot JSON config file | no | — |
| base-branch | Base branch for the PR | no | — |
| provider | LLM provider to use (openai|anthropic|gemini) | no | — |
| release-tag | Git ref (tag or HEAD) to generate changelog for | no | — |
| release-name | Display name for the version (without 'v') | no | — |
| release-body | Additional release notes body to merge in | no | — |
| language | Language for generated changelog prose | no | — |
| instructions | Additional changelog writing and grouping instructions | no | — |
| instructions-file | Path to a file with additional changelog instructions | no | — |
| npm-version | npm dist-tag or version range for @nyaomaru/changelog-bot | no | latest |
| minimum-package-age-days | Minimum age in days required for the npm package version resolved from npm-version | no | 0 |
| dry-run | If 'true', print updated changelog without writing or PR | no | — |
| dry-run-json-report | If 'true', print dry-run provider diagnostics as JSON | no | — |
| fail-on-llm-error | If 'true', fail instead of falling back when provider calls fail | no | — |
| require-provider | If 'true', fail when the selected provider API key is missing | no | — |
| no-ai | If 'true', skip all provider calls and use deterministic output | no | — |
| why | If 'true', extract short WHY notes from PR descriptions | no | — |
| why-max-prs | Maximum number of PRs to inspect for WHY extraction | no | — |
| why-max-chars-per-pr | Maximum preprocessed candidate characters sent per PR for WHY extraction | no | — |
| why-confidence | Minimum provider confidence for WHY notes (low|medium|high) | no | — |
| why-label | Label used for rendered WHY notes | no | — |
Outputs
no outputs