Claude Code Prompts: Best Templates and Developer Practices

You are Claude Code.

First, map this repo: list key directories, runtime, build/test commands, and where configs live.
Then propose a 30-minute onboarding plan for a new engineer.

Ask up to 5 clarifying questions if needed.

Run the test suite.

For each failing test:
1) explain the failure cause
2) propose the smallest safe code change
3) show how to prove the fix (tests/commands)

Prefer adding or adjusting tests over disabling them.
Output a step-by-step plan before editing files.

Refactor this module for readability without changing behavior.

First, write characterization tests that lock in current behavior.
Then refactor in small commits with explanations.
Call out any behavior changes you suspect.

Profile the slow path.
Identify bottlenecks, then propose 3 optimizations ranked by impact and risk.

For each:
- Files and functions: exact files/functions to change
- Tradeoffs: expected tradeoffs
- Benchmark plan: a benchmark plan (before/after)

Review changes for common web security issues:
- Authentication and authorization
- Injection (SQL/command/template, unsafe deserialization)
- SSRF (URL fetching, allowlists, egress)
- XSS (output encoding, sanitization)
- CSRF (tokens, same-site cookies)
- Secrets handling

Produce a prioritized list with severity, exploit scenario, and exact mitigation diff.

Implement: [feature].

Start by clarifying requirements, then design:
- Data model: tables/collections, keys, constraints, migrations
- API routes: endpoints, request/response shapes, auth, errors
- User interface: screens, states, validation, empty/loading behaviors

Propose a plan, then make changes.
Add tests and update docs. Keep diffs small and reviewable.

Summarize this change for a PR.

Include:
- Problem: what was broken or missing, who it impacted
- Solution: what you changed and why this approach
- Key files: main entry points a reviewer should inspect
- Tests run: exact commands and results
- Risks: what could go wrong in production
- Rollout plan: flags, staged rollout, deployment steps

Keep it skimmable.

Add logs, metrics, and traces for the critical path.

Define useful fields, avoid PII, include alerting thresholds.
Show exact insertion points and sample dashboards.

Upgrade dependency X to version Y safely.

Identify:
- Breaking changes: what upstream release notes say changed
- Affected paths: where those changes touch your code
- Staged plan: safest order of edits, tests, rollout
- Rollback steps: how to revert cleanly

Run tests and fix compile/runtime errors.

Turn this manual checklist into an idempotent script/CLI.

Requirements:
- Dry-run mode: show what will change without changing it
- Clear output: print what happened and what to do next
- Error handling: fail safely with actionable messages
- Docs: add a README or usage help

Prefer safe defaults.