v1.0.0

EvalCI v1.0.0

LLM quality gates for every PR — zero infra, 2-minute setup, works with any LLM provider.

Features

• Run @eval_case suites automatically on every pull request
• Block merge if quality drops below threshold
• Post formatted results table as a PR comment (score, cost, latency per case)
• Works with 30+ LLM providers via SynapseKit
• Zero infrastructure — runs entirely in GitHub Actions

Usage

- uses: SynapseKit/evalci@v1
  with:
    path: tests/evals
    threshold: "0.80"
  env:
    OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}

What's included

• Apache 2.0 license
• Issue templates (bug report, feature request)
• Discussion template
• PR template
• CONTRIBUTING.md, SECURITY.md, CHANGELOG.md

See README for full documentation.