# AI Agent Regression Testing

Compare baselines and candidates against pinned agent failures and datasets with replay evidence, release thresholds, and CI release gates.

Source: https://www.agentclash.dev/platform/agent-regression-testing
Markdown export: https://www.agentclash.dev/md/platform/agent-regression-testing

## Regression loop

- Capture failed production or evaluation traces.
- Curate them into repeatable cases.
- Run baseline and candidate agents on the same suite.
- Block regressions in CI with evidence-linked gates.