AI agents

Regression gates and release checks

• Feeding a Failed Sample Back Into the Tier's Prompt or Tool
• Turning a Retrieval Curve Into a Go/No-Go Release Rule
• A Decision Log for Waived Gates: Fields, Owners and Expiry
• A Gate That Blocks a Retrieval Change Before Release
• A Retrieval Regression Gate for Release
• A Waiver Log for Regression Fixtures a Team Chooses to Skip
• Canary Process for Prompt Changes After a Fixture Passes
• Choosing a Pass Threshold When Failures Are Asymmetric
• Regression Gates for Support Agents: Which Evals Block a Release
• The Lint Rule in CI: Blocking a Bad Generated Set
• Turning a Regression Fixture Into a Live Alert
• When a Retrieval Failure Should Block a Release
• When the Test Suite Misses: What a Failed Gate Costs
• When to Move a Failing Sample From Eval to Training
• When to Retire an Agent After a Run of Misses
• Where the Gate Runs: CI, Staging or Nightly
• Wiring the Citation Check Into CI