Act
AI Self-Healing CI/CD
When a test fails, an agent reads the failure, writes a fix, and opens a pull request — inside the same pipeline run that caught it.
What is actually interesting about it
- Parses the failure log, fetches the relevant source for context, generates a fix, commits to a branch and opens a PR. A human still merges — the loop ends at review, deliberately.
- Handles pytest, unittest and generic build errors rather than one framework.
- Prometheus metrics on fix rate, acceptance rate and time-to-repair, because an agent you cannot measure is an agent you cannot trust.
- The earlier, blunter ancestor of AISRE: same instinct, no capability boundary and no evaluator. Building it is what made the case for Atlas and KLRB.
Scope — what this is not
A working Jenkins integration. The fix quality is bounded by the model behind it — which is precisely the problem KLRB was built to measure.
Elsewhere in the system
Replaces the shared API key an AI agent uses with a scoped, attenuable token bound to its workload identity — verified offline, with no network hop on the authorization path.
An agentic incident-response pipeline: it detects a live service failure, investigates it with a three-tool orchestrator, and proposes a root cause — but cannot touch anything until a human approves.
A Kubernetes benchmark that measures whether an LLM actually read the cluster evidence before diagnosing an incident — or just guessed confidently from metadata.