michalzuch/Generate llms.txt
Generate llms.txt, llms-full.txt, and JSON-LD schema from your documentation for AI discoverability
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| content-dir | Directory containing HTML, Markdown, or framework source files to analyze | yes | — |
| site-url | Public base URL of the site (e.g. https://docs.example.com) | yes | — |
| site-name | Human-readable product or site name | yes | — |
| dist-dir | Built HTML output directory — enables JSON-LD injection into HTML files | no | "" |
| llm-base-url | OpenAI-compatible API base URL | no | http://localhost:11434/v1 |
| llm-model | Model identifier to use for analysis | no | qwen2.5:3b |
| llm-api-key | API key for the LLM provider (not required for Ollama) | no | ollama |
| ollama-setup | Install and start Ollama on the runner — disable when using an external LLM API | no | true |
| llm-concurrency | Number of concurrent LLM requests | no | 5 |
| commit | Commit and push generated files back to the branch | no | false |
| commit-message | Git commit message for the generated files | no | chore: regenerate llms.txt |
Outputs
no outputs