Regression checks
Compare recorded traces for citation, grounding and other quality regressions.
RAGOps · v2.0.2 · OPEN SOURCE
Compare recorded RAG and agent traces against an accepted baseline. Review regressions before deciding to ship.
Python 3.11+ · Offline core · v2.0.2
Baseline and candidate traces from the same scenario.
ragops-three-minute-demo
Accepted baselinebaseline.json
Candidate tracescandidate.json
Deterministic evaluation of recorded behavior.
| Metric | Baseline | Candidate |
|---|---|---|
| Citation coverage | 1.0 | 0.5 |
| Citation precision | 1.0 | 0.5 |
| Lexical groundedness | 1.0 | 0.6 |
Check gates against a versioned release policy.
Contract 2.0
Failed gatescandidate_release_gatecitation_coverage_regressioncitation_precision_regressiongroundedness_regressionOne canonical decision. Human owns final shipping.
⊘ BLOCK
Synthetic regression demo. Expected canonical decision: BLOCK. Bundle integrity is verified; model safety is not certified.
Python 3.11+. Run a disposable local demo.
pipx install ragops==2.0.2
ragops demo --profile engineer --output ./demo-output
ragops evidence verify --bundle ./demo-output/evidenceAlready have pipx? Use its isolated environment. Otherwise choose Python venv (macOS/Linux) or From source (macOS/Linux commands).
Recorded output from the source demo. This page does not run RAGOps or inspect your files.
The demo writes local reports to ./demo-output. Its BLOCK decision is intentional.
Read demo source →Compare recorded traces for citation, grounding and other quality regressions.
Keep gate thresholds and the canonical PASS/WARN/BLOCK decision inspectable.
Export reports and verify evidence bundle integrity without network calls.
Export HTML, JSON, Markdown, JUnit, SARIF and GitHub Summary. API/workbench and vendor adapters are optional. The core does not make network calls. Evidence verification checks integrity, not model safety.
Read documentation →