Practice Lesson 3 of 3

Team Operations

Overview

Integrating Claude Code into team workflows requires established conventions around pull requests, code reviews, and transparency regarding AI-assisted contributions. This lesson defines operational standards for each of these areas.

Pull Request Standards

A pull request should communicate scope, rationale, and verification status:

## Summary
Add rate limiting to the API to prevent abuse.

## Changes
- Added express-rate-limit middleware
- Configured 100 requests per 15 minutes per IP
- Added rate limit headers to responses

## Testing
- Verified rate limiting triggers after 100 requests
- Confirmed headers are present in responses

## Related
- Plan: docs/plans/20260509_rate_limiting.md
- Issue: #42

Note: Claude Code can generate PR descriptions from the changes made during a session. This reduces documentation overhead while maintaining quality.

Receiving Reviews

Effective review reception follows these principles:

  • Depersonalize feedback — Reviews evaluate code, not the author
  • Request rationale — "What specific concern does this change address?"
  • Leverage Claude Code for implementation — "The reviewer suggested X. Implement this approach."

Conducting Reviews

Effective review contributions are:

  • Specific — "This could cause a race condition on concurrent writes" identifies a concrete risk
  • Constructive — Include a suggested solution alongside the identified problem
  • Balanced — Acknowledge well-implemented sections as well as areas for improvement

AI Collaboration Standards

Transparency

When Claude Code generates significant portions of a contribution, this should be communicated to the team. Context about the generation process informs review priorities.

Ownership

The submitter bears full responsibility for all committed code, regardless of its generation method. Every change must be reviewed and understood prior to submission.

Knowledge sharing

Effective prompts, workflows, and CLAUDE.md configurations benefit the entire team. Documenting these practices in shared resources multiplies their value.

Requirement: Code that cannot be explained and maintained by the submitter should not be committed. Comprehension is a prerequisite for ownership.

Key Takeaways

  • PRs document scope, rationale, and verification method
  • Code reviews evaluate implementation quality, not the author
  • AI assistance should be transparent to the team
  • Comprehension is a prerequisite for code submission