controlnet/auto-llm-changelog
Generate a Markdown changelog from git history and diff using an OpenAI-compatible Chat Completions API.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| api_endpoint | OpenAI-compatible Chat Completions endpoint URL. | yes | — |
| api_key | API key used as Bearer token. | yes | — |
| model | Model name passed to the Chat Completions API. | yes | — |
| current_ref_override | Explicit ref to use as current_ref. | no | "" |
| current_version_override | Explicit version to use as current_version. | no | "" |
| previous_tag_override | Explicit previous semver tag to use. | no | "" |
| temperature | Sampling temperature sent to the API. | no | 0.2 |
| max_diff_bytes | Maximum bytes for the aggregate diff section included in the prompt. | no | 200000 |
| system_prompt | Override the default system prompt. | no | "" |
| fetch_remote_refs | Whether to fetch tags and remote refs before resolving ranges. | no | true |
| include_commit_patches | Whether to include per-commit patches in the prompt, subject to size limits. | no | false |
| debug | Enable verbose debug logging. | no | false |
| max_prompt_bytes | Soft cap for the assembled user prompt body in bytes before optional sections are trimmed. | no | 300000 |
Outputs
| name | description |
|---|---|
| markdown | Generated Markdown changelog. |
| current_ref | Resolved current_ref used for git comparison. |
| version | Resolved current_version, or empty string when unavailable. |
| previous_tag | Resolved previous semver tag. |
| compare_range | Git compare range in previous_tag..current_ref form. |