riccardomerenda/PromptScore
Lint your prompts before they reach a model. Static analysis for LLM prompts.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit May 2, 2026
- License
- MIT
Pinned Snippet
uses: riccardomerenda/promptscore@abff136e0d6b5c7a7bb921169b82c144ce8e1f72 # v0.4.8tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| inputs | Files, directories, or globs to analyze. Whitespace-separated. Defaults to "prompts/". | no | prompts/ |
| model | Profile name. Built-in: _base, claude, gpt. Custom profiles supported via project config. | no | _base |
| format | Report format: text, markdown, or json. Markdown is also appended to the GitHub Actions job summary. | no | text |
| fail-on | Severity threshold that fails the action: error, warning, info, or none. | no | error |
| config | Path to a PromptScore config file. Defaults to project discovery. | no | "" |
| rules | Comma-separated rule IDs to include. Empty means all rules from the active profile. | no | "" |
| include-llm | Run LLM-backed rules. Requires the relevant API key in the runner environment. | no | false |
| cli-version | Version of @promptscore/cli to install (npm semver tag or version). Defaults to latest. | no | latest |
| node-version | Node.js version to use on the runner. | no | 20 |
Outputs
no outputs