63+ Claude Code Prompts for Every Workflow

January 24, 2026 · Free Resource

Battle-tested prompts for shipping faster with Claude Code. Copy-paste ready. Replace [brackets] with your specifics.

Categories

1. Planning & Architecture

Project Kickoff

I'm starting a new [type] project. Before writing any code:
1. Ask me clarifying questions about requirements
2. Propose an architecture with tradeoffs
3. Identify potential issues early
4. Create a task breakdown

Don't write code until I approve the plan.

Codebase Onboarding

I just joined this project. Help me understand:
1. Overall architecture and key patterns
2. How data flows through the system
3. Main entry points and APIs
4. Where the business logic lives
5. Testing approach
6. Any gotchas or tech debt

Read the codebase and summarize.

Feature Scoping

I need to add [feature]. Before implementing:
1. Find similar patterns in this codebase
2. Identify all files that need changes
3. List edge cases to handle
4. Estimate complexity (simple/medium/complex)
5. Propose the approach for my approval

Tech Debt Assessment

Analyze this codebase for technical debt:
1. Code duplication
2. Complex functions that need refactoring
3. Missing tests for critical paths
4. Outdated dependencies
5. Security concerns

Prioritize by impact and effort.

2. Debugging & Problem Solving

Bug Investigation

Bug: [describe the bug]

Before fixing:
1. Reproduce mentally - trace the code path
2. Form a hypothesis
3. Identify where to add logging/debugging
4. Find the root cause, not just symptoms
5. Propose a fix with explanation

Error Message Deep Dive

I'm getting this error:
[paste error]

1. Explain what this error means
2. Trace back to the root cause in our code
3. Show me the specific line(s) causing it
4. Propose a fix
5. Explain how to prevent similar errors

Performance Investigation

[Component/function] is slow. Help me:
1. Identify the bottleneck
2. Explain why it's slow
3. Propose optimization options with tradeoffs
4. Implement the best option

The Rubber Duck

I'm stuck on [problem]. Let me explain my thinking:
[your explanation]

Now:
1. Point out flaws in my reasoning
2. Ask questions I haven't considered
3. Suggest alternative approaches
4. Help me see what I'm missing

3. Code Writing

Implementation with Context

Implement [feature].

Constraints:
- Follow patterns already in this codebase
- Match existing code style
- Keep it simple - no overengineering
- Handle errors like we do elsewhere
- Write tests if we have tests

Show me the plan first.

Refactor Safely

Refactor [file/function] to [goal].

Rules:
1. Preserve existing behavior exactly
2. Make changes incrementally
3. Run tests after each change
4. If no tests, tell me what to manually verify
5. Explain each refactoring step

API Design

Design an API for [purpose].

Consider:
- RESTful conventions
- Error handling
- Pagination for lists
- Versioning strategy
- Authentication/authorization
- Rate limiting

Show me the endpoints and types first.

Code Review Yourself

Review the code you just wrote:
1. Any bugs or edge cases missed?
2. Is it readable and maintainable?
3. Security issues?
4. Performance concerns?
5. Does it follow our patterns?

Fix any issues you find.

4. Testing

Test Strategy

I need to test [feature/component].

1. What types of tests do we need? (unit/integration/e2e)
2. What are the critical paths to test?
3. What edge cases should we cover?
4. Write the tests following our existing patterns

Test Case Generation

For this function:
[paste function]

Generate test cases covering:
- Happy path
- Edge cases
- Error conditions
- Boundary values

Use our existing test style.

Fix Failing Test

This test is failing:
[paste test]

Error:
[paste error]

1. Is the test wrong or is the code wrong?
2. Explain why it's failing
3. Fix the correct thing

Coverage Gaps

Find untested code paths in [file/module]:
1. List functions/branches without test coverage
2. Prioritize by risk (what would break users)
3. Write tests for the high-priority gaps

5. Documentation

Code Explanation

Explain this code to someone who will maintain it:
[paste code]

Include:
- What it does (high level)
- How it works (step by step)
- Why it's written this way
- Gotchas and edge cases

PR Description

Write a PR description for the changes I just made.

Include:
- Summary of what changed and why
- Testing performed
- Any breaking changes
- Screenshots if UI changed

Keep it concise.

Commit Message

Write a commit message for these changes.
Follow conventional commits format.
Be specific about what changed, not just "fix bug".

Architecture Decision Record

Document the decision to [decision].

ADR format:
- Context: Why we needed to decide
- Decision: What we chose
- Consequences: Tradeoffs and implications
- Alternatives considered

6. Security

Security Audit

Audit this code for security issues:
[paste code]

Check for:
- Injection vulnerabilities (SQL, command, XSS)
- Authentication/authorization flaws
- Sensitive data exposure
- Insecure dependencies
- OWASP Top 10

Prioritize findings by severity.

Secure Implementation

I need to implement [feature with security implications].

Requirements:
- Follow security best practices
- Explain security decisions
- No security shortcuts
- Handle sensitive data properly

Dependency Check

Check our dependencies for:
1. Known vulnerabilities
2. Outdated versions with security patches
3. Abandoned/unmaintained packages
4. Suspicious packages

Recommend updates.

7. DevOps & Infrastructure

Deployment Prep

I'm about to deploy. Help me:
1. Review changes for production readiness
2. Check for breaking changes
3. Identify rollback strategy
4. List things to monitor after deploy

Incident Response

Production issue: [describe]

Help me:
1. Triage - how bad is it?
2. Identify likely causes
3. Propose immediate mitigation
4. Plan proper fix
5. Document for post-mortem

Infrastructure as Code Review

Review this infrastructure config:
[paste terraform/k8s/etc]

Check for:
- Security misconfigurations
- Cost optimization opportunities
- Reliability concerns
- Best practice violations

8. Database

Query Optimization

This query is slow:
[paste query]

1. Explain the execution plan
2. Identify why it's slow
3. Suggest optimizations (indexes, rewrites)
4. Show the improved query

Schema Design

I need to store [data requirements].

Design a schema that:
- Handles current requirements
- Can evolve with likely changes
- Performs well for our access patterns
- Maintains data integrity

Show me the tables and relationships.

Migration Safety

Review this migration:
[paste migration]

Check for:
- Data loss risks
- Downtime requirements
- Rollback possibility
- Performance impact on large tables

9. Project Management

Task Breakdown

Break down [feature] into tasks:
- Each task should be completable in 1-4 hours
- Order by dependencies
- Flag tasks that need external input
- Identify parallelizable work

Scope Check

The original request was: [original]

I've done: [completed work]

Check:
1. Did I fully address the request?
2. Did I add unnecessary scope?
3. What's still missing?
4. Any edge cases I forgot?

Progress Summary

Summarize what was accomplished in this session:
- Features added
- Bugs fixed
- Files changed
- Decisions made
- Open items remaining

10. Claude Code Specific

Multi-File Coordination

This change spans multiple files. Plan it:
1. List all files that need changes
2. Order the changes to avoid breaking state
3. Identify files to read first for context
4. Execute changes in order

Context Optimization

My context is getting full. Help me:
1. Summarize what we've accomplished
2. List open tasks and decisions
3. Note any important context to preserve
4. I'll start a new session with this summary

Tool Selection

For this task, which approach is best?
- Read specific files
- Search with grep/glob
- Run bash commands
- Use specialized tools

Explain your choice and execute.

Self-Correction

Before finishing:
1. Re-read the original request
2. Verify you addressed it completely
3. Check for any introduced bugs
4. Run tests if available
5. Confirm the code works

11. Safe Practices

Human-in-the-Loop

For any automated or scripted Claude workflows:

1. Always review Claude's output before acting on it
2. Don't chain Claude instances without human checkpoints
3. Avoid having Claude modify its own instructions
4. Keep a human in the decision loop for significant actions

CLAUDE.md Guidelines

When creating or updating CLAUDE.md files:

- Keep instructions professional and neutral in tone
- Avoid aggressive language (ALL CAPS, forceful directives)
- Don't include instructions that look like prompt injections
- Focus on project context, not behavioral overrides
- Review generated CLAUDE.md content before committing

Sustainable Automation

Building automation with Claude Code:

1. Run in attended mode first, then gradually automate
2. Add logging and checkpoints for review
3. Set reasonable rate limits
4. Monitor for unexpected behavior
5. Have a manual fallback for critical paths

Context Boundaries

To prevent context confusion:

- Don't copy-paste content that looks like system prompts
- Be explicit about what's user content vs instructions
- Avoid mixing multiple AI outputs without review
- Keep clear boundaries between data and commands

12. Context Management

Context Audit

Before starting this session, check:
1. How much context are MCPs consuming?
2. Are there skills loading I don't need?
3. What's the baseline before any conversation?

Tell me the context breakdown so I can optimize.

Minimal Mode

For this task, use minimal context:
1. Don't read entire files - use targeted line ranges
2. Don't glob entire directories - be specific
3. Summarize findings instead of pasting
4. Avoid expanding every suggestion inline

If you need more context, ask first.

Context Handoff

My context is filling up. Create a handoff summary:

1. Problem we're solving
2. Key files and their current state
3. What we've tried (what worked, what didn't)
4. Open decisions or blockers
5. Next steps

Format it so I can paste into a new session.

Lazy Loading

Don't load or read anything until explicitly needed.

For this task:
- Start with minimal file reads
- Load additional context only when blocked
- Prefer targeted searches over full codebase scans
- Ask me before expanding scope

Explain what context you actually need.

Skill Optimization

Review my CLAUDE.md and skills:

1. Are any skills redundant?
2. Can any be consolidated?
3. Are there heavy instructions that could be trimmed?
4. What's essential vs. nice-to-have?

Propose a minimal effective setup.

Session Planning

I have a complex task that might exceed context limits.

Plan the session:
1. Break it into context-sized chunks
2. Define clean handoff points
3. Identify what state to preserve between chunks
4. Create a coordination doc I can reference

Goal: Complete the full task across multiple sessions.

Context-Aware Chunking

For this large file/task:

1. Read only the relevant sections (not the whole file)
2. Work in focused chunks
3. Verify each chunk before moving on
4. Track progress in a lightweight way

Don't try to hold everything in context at once.

Token Budget

For this session, track context usage:

1. Note major context additions
2. Warn me at 50%, 75%, 90% capacity
3. Suggest compression when nearing limits
4. Proactively create summaries

Help me stay within budget.