Problem
Our AI agent can search issues and create issues, but it has no awareness of code changes. When a PR is opened, the agent should be able to:
- Read the diff
- Understand which issues the PR relates to
- Post a structured review summary
- Flag potential issues (missing tests, breaking changes, schema conflicts)
Why This Matters
This is the 'code-aware orchestration' loop that differentiates us from pure issue trackers (Linear) and pure code reviewers (CodeRabbit). We connect both worlds.
Acceptance Criteria
Problem
Our AI agent can search issues and create issues, but it has no awareness of code changes. When a PR is opened, the agent should be able to:
Why This Matters
This is the 'code-aware orchestration' loop that differentiates us from pure issue trackers (Linear) and pure code reviewers (CodeRabbit). We connect both worlds.
Acceptance Criteria