48 lines (32 loc) · 1.89 KB

CLAUDE.md

Workflow Orchestration

1. Plan Mode Default

Enter plan mode for ANY non-trivial task (3+ steps or architectural decisions)
If something goes sideways, STOP and re-plan immediately - don't keep pushing
Use plan mode for verification steps, not just building
Write detailed specs upfront to reduce ambiguity

2. Subagent Strategy

Use subagents liberally to keep main context window clean
Offload research, exploration, and parallel analysis to subagents
For complex problems, throw more compute at it via subagents
One task per subagent for focused execution

3. Self-Improvement Loop

After ANY correction from the user: update the closest relevant CLAUDE.md file
Write rules for yourself that prevent the same mistake

4. Verification Before Done

Never mark a task complete without proving it works
Diff behavior between main and your changes when relevant
Ask yourself: "Would a staff engineer approve this?"
Run relevant tests only, check logs, demonstrate correctness

5. Demand Elegance (Balanced)

For non-trivial changes: pause and ask "is there a more elegant way?"
If a fix feels hacky: "Knowing everything I know now, implement the elegant solution"
Skip this for simple, obvious fixes - don't over-engineer
Challenge your own work before presenting it

Core Principles

Minimal Impact: Changes should only touch what's necessary. Avoid introducing bugs.

Working Constraints

Worktrees must use a full yarn install in the worktree. Do not symlink node_modules or packages back to another checkout.
E2E and packaged-binary integration tests do not run in the sandbox. When they are needed, provide the exact commands and ask the user to run them.
Before committing or pushing, yarn check-ts, yarn lint, and relevant unit tests must pass locally. CI currently does not block expensive e2e jobs when these fail.

@AGENTS.md