omp packages

pi-commit-impact1.0.0

Pi extension that estimates production regression risk for commits using Jev HTTP plus local heuristics.

marcelblijleven Published 26 days ago ~171 downloads/mo

Untested on omp

This is a pi package. No one has reported whether it works on omp yet. If you try it, add your result below.

Did it work for you?

Result

One report per person. A new report replaces your old one.

Reports 0

No reports yet.

Readme

pi-commit-impact

Pi extension that estimates the production regression risk of a commit/range using deterministic local heuristics plus optional TypeSafe.ai Jev HTTP analysis.

Run locally

npm install
pi --extension .

Then use the Pi slash command:

/commit-impact
/commit-impact HEAD --format markdown --fail-on high
/commit-impact --last 3 --format markdown
/commit-impact main..HEAD --format json --config pi-commit-impact.config.json

The extension also registers a model-callable commit_impact tool with the same core options. The tool supports lastCommits, so Pi can directly answer natural requests such as “what's the risk score of the last 3 commits?” without translating that into a git range itself.

Use from other agents

Other coding agents can call the CLI through their shell tool:

npx pi-commit-impact --last 3 --format json
commit-impact HEAD --format markdown --fail-on high
commit-impact main..HEAD --config pi-commit-impact.config.json

When developing from this repo, run the local executable directly:

./commit-impact.js --last 3 --format json

The CLI exits with code 1 when --fail-on is supplied and the analyzed risk tier meets or exceeds the threshold, making it suitable for CI or agent guardrails.

Configuration

Create pi-commit-impact.config.json at the repository root. See pi-commit-impact.config.example.json.

Secrets are not stored in config. Configure the TypeSafe System One endpoint (https://api.typesafe.ai/v1/systemone) in config and set jev.tokenEnv to the name of an environment variable containing the API key, commonly TYPESAFE_API_KEY.

The Jev request uses typed System One questions: a production-regression score, a critical-behavior-change noul, and a validation-focus choice. If jev is absent from config, the extension silently uses heuristic-only analysis. If jev is configured but invalid or the HTTP call fails, it returns heuristic-only analysis with a warning.

Signals

The MVP scores:

  • conventional commit type and configurable keywords
  • changed file count
  • line churn
  • critical path globs
  • file types/extensions
  • whether tests changed
  • Jev HTTP score/tier/evidence when configured

Output is Markdown or JSON and does not require custom TUI rendering.