Skip to content
Back to search

claude-code-prompts

Independently authored prompt templates for AI coding agents — system prompts, tool prompts, agent delegation, memory management, and multi-agent coordination. Informed by studying Claude Code.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/repowise-dev--claude-code-prompts/badge.svg)](/listings/repowise-dev--claude-code-prompts)

HTML

<a href="/listings/repowise-dev--claude-code-prompts"><img src="/api/listings/repowise-dev--claude-code-prompts/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

1. Browse the [pattern files](#prompt-engineering-patterns) to understand how production coding agents structure their prompts. 2. Copy any [complete prompt](#claude-code-system-prompt-1) into your own agent configuration. 3. Replace `{{PLACEHOLDER}}` values with your stack, tool names, and risk policy. 4. For Cursor IDE users, drop `skills/` into `~/.cursor/skills-cursor/` for instant integration. 5. Add codebase context with [RepoWise](https://github.com/repowise-dev/repowise) so your agent actually understands the project it's working in.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 176 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Claude Code Prompts — Independently Authored Prompt Collection
  2. Independently written prompt templates for building AI coding agents, inspired by patterns observed in Claude Code codebase. System prompts, tool prompts, agent prompts, memory management, and multi-agent coordination.
  3. ## Check out [RepoWise](https://github.com/repowise-dev/repowise)
  4. <p align="center">
  5. <a href="https://github.com/repowise-dev/repowise">
  6. <img src="./assets/repowise-logo.png" alt="RepoWise" width="140" />
  7. </a>
  8. </p>
  9. **Prompts tell your agent how to behave. [RepoWise](https://github.com/repowise-dev/repowise) tells it what the codebase actually does.**
  10. Open-source codebase intelligence engine. Auto-generated architecture wikis, dependency graphs, ownership maps, dead code detection, and 9 MCP tools your AI editor can query in real time.
  11. **[GitHub](https://github.com/repowise-dev/repowise)** &nbsp; | &nbsp; [repowise.dev](https://repowise.dev)
  12. Scroll down for Claude Code's prompt patterns. 👇
  13. ---
  14. <!-- GitHub repo description: Independently authored prompt templates for AI coding agents — system prompts, tool prompts, agent delegation, memory management, and multi-agent coordination. Informed by observing how Claude Code works in practice. -->
  15. <!-- Suggested GitHub topics: claude-code, claude-code-prompts, claude-prompts, system-prompt, claude-code-system-prompt, prompt-engineering, anthropic-claude, ai-coding-agent, claude-agent, llm-prompts, ai-prompts, cursor-skills, multi-agent, coding-assistant, ai-tools, prompt-templates -->
  16. ## Claude Code System Prompts
  17. We studied how Claude Code behaves in practice -- how it structures safety rules, routes tools, delegates to subagents, manages memory -- and wrote our own prompt collection from scratch implementing the same patterns.
  18. **Every prompt in this repo is independently authored.** We observed what patterns make a coding agent reliable and wrote our own versions. From the team behind [RepoWise](https://github.com/repowise-dev/repowise).
  19. Not affiliated with Anthropic. See [DISCLAIMER.md](./DISCLAIMER.md).
  20. > **Note**
  21. > ⭐ Star this repository to if you find the collection useful.
  22. ## What's Inside
  23. | Category | Files | What You Get |
  24. |----------|-------|--------------|
  25. | [System prompt](#claude-code-system-prompt-1) | 1 | Agent identity, safety rules, code style, tool routing, output format |
  26. | [Tool prompts](#claude-code-tool-prompts) | 11 | Shell, file read/edit/write, grep, glob, web search/fetch, agent launcher, ask user, plan mode |
  27. | [Agent prompts](#claude-code-agent-prompts) | 5 | General purpose, code explorer, solution architect, verification specialist, documentation guide |
  28. | [Memory prompts](#claude-code-memory-prompts) | 4 | Conversation summarization, session notes, memory extraction, memory consolidation |
  29. | [Coordinator prompt](#multi-agent-coordinator) | 1 | Multi-worker orchestration with synthesis, delegation, and verification workflow |
  30. | [Utility prompts](#claude-code-utility-prompts) | 4 | Session titles, tool summaries, away recaps, next-action suggestions |
  31. | [Pattern analyses](#prompt-engineering-patterns) | 9 | Commentary on each pattern with reusable templates |
  32. | [Cursor skills](#cursor-skills) | 3 | Drop-in skills for coding standards, verification, and prompt design |
  33. ## Quick Start
  34. 1. Browse the [pattern files](#prompt-engineering-patterns) to understand how production coding agents structure their prompts.
  35. 2. Copy any [complete prompt](#claude-code-system-prompt-1) into your own agent configuration.
  36. 3. Replace `{{PLACEHOLDER}}` values with your stack, tool names, and risk policy.
  37. 4. For Cursor IDE users, drop `skills/` into `~/.cursor/skills-cursor/` for instant integration.
  38. 5. Add codebase context with [RepoWise](https://github.com/repowise-dev/repowise) so your agent actually understands the project it's working in.
  39. ## Prompt Engineering Patterns
  40. Patterns observed in how Claude Code structures its prompts — with analysis and independently written templates.
  41. - [System Prompt Architecture](./patterns/01-system-prompt-architecture.md) — layered identity, safety, and behavior
  42. - [Core Behavioral Rules](./patterns/02-core-behavioral-rules.md) — anti-over-engineering, minimal changes
  43. - [Safety and Risk Assessment](./patterns/03-safety-and-risk-assessment.md) — reversibility tiers, destructive action gates
  44. - [Tool-Specific Instructions](./patterns/04-tool-specific-instructions.md) — per-tool routing and constraints
  45. - [Agent Delegation](./patterns/05-agent-delegation.md) — when and how to spawn subagents
  46. - [Verification and Testing](./patterns/06-verification-and-testing.md) — adversarial verification, anti-rationalization
  47. - [Memory and Context](./patterns/07-memory-and-context.md) — summarization, session notes, memory extraction
  48. - [Multi-Agent Coordination](./patterns/08-multi-agent-coordination.md) — coordinator pattern, worker synthesis
  49. - [Auxiliary Prompts](./patterns/09-auxiliary-prompts.md) — utility prompts for titles, recaps, suggestions
  50. ## Claude Code System Prompt
  51. The complete system prompt covering identity, permissions, task execution, code style, safety, tool routing, tone, and output efficiency.
  52. - [System Prompt](./complete-prompts/system-prompt.md) — the main agent identity and behavioral rules
  53. - [Coordinator Prompt](./complete-prompts/coordinator-prompt.md) — multi-agent orchestration mode
  54. ## Claude Code Tool Prompts
  55. Prompt templates for each tool a coding agent uses to interact with the filesystem, shell, web, and user.
  56. - [Shell Execution](./complete-prompts/tool-prompts/shell-execution.md) — bash command execution with sandbox, git safety, commit/PR workflows
  57. - [File Read](./complete-prompts/tool-prompts/file-read.md) — reading files, images, PDFs, notebooks
  58. - [File Edit](./complete-prompts/tool-prompts/file-edit.md) — exact string replacement with uniqueness constraints
  59. - [File Write](./complete-prompts/tool-prompts/file-write.md) — creating new files with read-first safety
  60. - [Search Grep](./complete-prompts/tool-prompts/search-grep.md) — ripgrep-powered content search
  61. - [Search Glob](./complete-prompts/tool-prompts/search-glob.md) — fast file pattern matching
  62. - [Web Search](./complete-prompts/tool-prompts/web-search.md) — web search with mandatory source citations
  63. - [Web Fetch](./complete-prompts/tool-prompts/web-fetch.md) — URL fetching with AI-powered extraction
  64. - [Agent Launcher](./complete-prompts/tool-prompts/task-management.md) — spawning autonomous subagents
  65. - [Ask User](./complete-prompts/tool-prompts/ask-user.md) — structured multiple-choice questions
  66. - [Plan Mode](./complete-prompts/tool-prompts/plan-mode.md) — entering planning mode before implementation
  67. ## Claude Code Agent Prompts
  68. Prompt templates for specialized subagents — each designed for a different task type.
  69. - [General Purpose Agent](./complete-prompts/agent-prompts/general-purpose.md) — research, search, multi-step tasks
  70. - [Code Explorer Agent](./complete-prompts/agent-prompts/code-explorer.md) — read-only codebase navigation
  71. - [Solution Architect Agent](./complete-prompts/agent-prompts/solution-architect.md) — design and planning
  72. - [Verification Specialist Agent](./complete-prompts/agent-prompts/verification-specialist.md) — adversarial testing with PASS/FAIL/PARTIAL verdicts
  73. - [Documentation Guide Agent](./complete-prompts/agent-prompts/documentation-guide.md) — documentation tasks
  74. ## Claude Code Memory Prompts
  75. How to manage context across long sessions — summarization, session notes, and persistent memory.
  76. - [Conversation Summary](./complete-prompts/memory-prompts/conversation-summary.md) — 9-section context compression with no-tools constraint
  77. - [Session Notes](./complete-prompts/memory-prompts/session-notes.md) — structured session state tracking
  78. - [Memory Extraction](./complete-prompts/memory-prompts/memory-extraction.md) — persistent memory from recent messages
  79. - [Memory Consolidation](./complete-prompts/memory-prompts/memory-consolidation.md) — periodic memory cleanup and merging
  80. ## Claude Code Utility Prompts
  81. Small helper prompts for session management and user experience.
  82. - [Session Title](./complete-prompts/utility-prompts/session-title.md) — 3-7 word session title generation
  83. - [Tool Summary](./complete-prompts/utility-prompts/tool-summary.md) — 30-character tool action summaries
  84. - [Away Recap](./complete-prompts/utility-prompts/away-recap.md) — "while you were away" session recaps
  85. - [Next Action Suggestion](./complete-prompts/utility-prompts/next-action-suggestion.md) — suggested follow-up actions
  86. ## Cursor Skills
  87. Drop-in skills for Cursor IDE that implement these patterns directly.
  88. - [Skills Overview](./skills/README.md)
  89. - [Coding Agent Standards](./skills/coding-agent-standards/SKILL.md) — behavioral defaults for coding agents
  90. - [Verification Agent](./skills/verification-agent/SKILL.md) — adversarial verification workflow ([strategies](./skills/verification-agent/strategies.md))
  91. - [Prompt Architect](./skills/prompt-architect/SKILL.md) — prompt design methodology ([reference](./skills/prompt-architect/reference.md))
  92. ## Methodology
  93. We studied how Claude Code behaves in practice and identified the behavioral rules, safety patterns, tool-routing strategies, and multi-agent coordination approaches that make it effective. We then independently authored every prompt in this repository from scratch, implementing the same patterns in our own words.
  94. All content is original.
  95. ## Pair With RepoWise for Full-Stack Agent Intelligence
  96. The prompts in this repo handle agent behavior: safety, tool routing, delegation, verification. But behavior without context is like a senior engineer who's never seen the codebase. [RepoWise](https://github.com/repowise-dev/repowise) fills that gap.
  97. | What you get | How it helps your agent |
  98. |---|---|
  99. | Architecture wiki | Agent understands module boundaries before making changes |
  100. | Dependency graphs | Agent traces impact of changes across packages |
  101. | Ownership tracking | Agent knows which team/person owns each module |
  102. | Dead code detection | Agent avoids modifying unused code paths |
  103. | 9 MCP tools | Agent queries project knowledge directly from Cursor, Claude Code, or any MCP-compatible editor |
  104. Works with any language, any repo size.
  105. **[GitHub](https://github.com/repowise-dev/repowise)** | [repowise.dev](https://repowise.dev)
  106. ## Legal
  107. - [DISCLAIMER.md](./DISCLAIMER.md) — independent authorship, nominative fair use, non-affiliation, DMCA policy.
  108. - [LICENSE](./LICENSE) — MIT for all original content.
  109. - [CHANGELOG.md](./CHANGELOG.md) — version history.
  110. "Claude Code" is a trademark of Anthropic, used here under nominative fair use to describe the subject matter. This project is independent and not affiliated with Anthropic.
  111. ## Contributing
  112. - All additions must be original writing. No verbatim vendor prompt text.
  113. - Prefer concise, implementation-ready language over theory.
  114. - Every new pattern file must include at least one concrete template.
  115. - Follow section conventions: `Approach / Output / Constraints` for agents, `Include / Format / Constraints` for memory, `Rules / Format` for utilities.
  116. ## Star History
  117. If this helped you build better AI agents, star the repo so others can find it.
  118. ---
  119. **Built by [RepoWise](https://github.com/repowise-dev/repowise)** — open-source codebase intelligence for AI agents ([repowise.dev](https://repowise.dev)).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.