Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT 5.5 Thinking, GPT 5.5 Instant, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly.
This repository collects and archives extracted system prompts from major AI assistants, updated regularly as new model versions ship. For Anthropic products specifically, it includes prompts for Claude Fable 5, Claude Opus 4.8, multiple Claude Code variants (with separate files for the glob tool, grep tool, deferred tools, and bundled skills), Claude Cowork, Claude Design, and integrations such as Claude for Excel, Word, PowerPoint, and iOS. Beyond Anthropic, the repository covers OpenAI's GPT-5.5 Thinking, GPT-5.5 Instant, and Codex; Google's Gemini 3.5 Flash and Antigravity CLI; xAI's Grok; and tools like VS Code Copilot Agent, Cursor, and Perplexity. Where prompts change between versions, the repository links to diff views, such as the documented change from Claude Opus 4.8 to Claude Fable 5. The primary audience is prompt engineers, AI researchers, and developers who want to understand how these systems are instructed, reverse-engineer behavioral constraints, or benchmark prompt changes across model releases. The repository received coverage in The Washington Post in May 2026.
Repository collecting extracted/leaked system prompts from major AI chatbots and coding assistants.
- ✓Open-source license (CC0-1.0)
- ✓Actively maintained (<30d)
- ✓Healthy fork ratio
- ✓Clear description
- ✓Topics declared
- ✓Mature repo (>1y old)
git clone https://github.com/asgeirtj/system_prompts_leaksResumen de Tools
> **As seen in The Washington Post:** [See the hidden rules behind AI. Then use them to rewrite this article.](https://wapo.st/49t4gSb) (May 11, 2026) # System Prompts Leaks The purpose of this repo is to document the System Prompt instructions for all the AI chatbots out there - Claude, ChatGPT, Gemini etc. <img alt="ChatGPT leaking its system prompt after being asked to repeat all of the above" src="https://github.com/user-attachments/assets/0037a6c5-2ae4-4d34-8be0-0d679773172b" /> [](https://github.com/asgeirtj/system_prompts_leaks)  [](http://makeapullrequest.com) > 🆕 **[Diff: Claude Opus 4.8 → Claude Fable 5](https://www.diffchecker.com/QJn9jFNk/)** — see exactly what changed in the claude.ai system prompt for Anthropic's newest model ## Recently Updated | What | Date | Link | |------|------|------| | **Claude Fable 5** | June 9, 2026 | [System prompt](Anthropic/claude-fable-5.md) · [Diff vs Opus 4.8](https://www.diffchecker.com/QJn9jFNk/) | | **Claude Opus 4.8** | June 9, 2026 | [System prompt](Anthropic/claude-opus-4.8.md) · [Official](Anthropic/Official/2026-05-28-claude-opus-4.8.md) | | **Claude Code Glob & Grep tools** | June 9, 2026 | [Glob](Anthropic/Claude%20Code/glob-tool.md) · [Grep](Anthropic/Claude%20Code/grep-tool.md) | | **Claude Code (Opus 4.8)** | May 28, 2026 | [System prompt](Anthropic/Claude%20Code/claude-code-opus-4.8.md) | | **Claude Code & Cowork** | May 28, 2026 | [Claude Code](Anthropic/Claude%20Code/claude-code-opus-4.6.md) · [Cowork](Anthropic/claude-cowork.md) · [Cowork Dispatch](Anthropic/claude-cowork-dispatch.md) | | **GPT-5.5** | May 24, 2026 | [Thinking](OpenAI/gpt-5.5-thinking.md) · [Instant](OpenAI/gpt-5.5-instant.md) · [API](OpenAI/gpt-5.5-api.md) · [Pro API](OpenAI/gpt-5.5-pro-api.md) | | **Perplexity Computer** | May 21, 2026 | [System prompt](Perplexity/perplexity-computer.md) | | **VS Code Copilot Agent** | May 21, 2026 | [System prompt](Microsoft/vscode-copilot-agent.md) | | **Docker Gordon AI** | May 21, 2026 | [System prompt](Misc/docker-gordon-ai.md) | | **Gemini 3.5 Flash** | May 20, 2026 | [System prompt](Google/gemini-3.5-flash.md) · [AI Studio](Google/gemini-3.5-flash-ai-studio.md) · [Tools](Google/gemini-3.5-flash-tools.json) | | **Antigravity CLI** | May 20, 2026 | [System prompt](Google/antigravity-cli.md) | | **Zed AI** | May 16, 2026 | [System prompt](Misc/zed.md) | | **Grok Expert** | May 11, 2026 | [System prompt](xAI/grok-expert.md) | ---  ## Anthropic — Claude | Model | Prompt | |-------|--------| | **Claude Fable 5** | [**System prompt**](Anthropic/claude-fable-5.md) | | **Claude Opus 4.8** | [**System prompt**](Anthropic/claude-opus-4.8.md) | | **Claude Code (Opus 4.8)** | [**System prompt**](Anthropic/Claude%20Code/claude-code-opus-4.8.md) | | **Claude Opus 4.7** | [**System prompt**](Anthropic/claude-opus-4.7.md) | | **Claude Code (Opus 4.6)** | [**System prompt**](Anthropic/Claude%20Code/claude-code-opus-4.6.md) | | **Claude Opus 4.6** | [**System prompt**](Anthropic/claude-opus-4.6.md) | | **Claude Sonnet 4.6** | [**System prompt**](Anthropic/claude-sonnet-4.6.md) | | Claude.ai | [Anthropic Reminders](Anthropic/anthropic_reminders.md) | <details><summary>Integrations, official prompts & older versions</summary> | | | |--|--| | Integrations | [Cowork](Anthropic/claude-cowork.md) · [Cowork Dispatch](Anthropic/claude-cowork-dispatch.md) · [Desktop Code](Anthropic/claude-desktop-code.md) · [Design](Anthropic/claude-design.md) · [Mobile iOS](Anthropic/claude-mobile-ios.md) · [In Chrome](Anthropic/claude-in-chrome.md) · [For Excel](Anthropic/claude-for-excel.md) · [For Word](Anthropic/claude-for-word.md) · [In PowerPoint](Anthropic/claude-in-powerpoint.md) · [Default Styles](Anthropic/default-styles.md) | | Claude Code extras | [Glob tool](Anthropic/Claude%20Code/glob-tool.md) · [Grep tool](Anthropic/Claude%20Code/grep-tool.md) · [Deferred tools](Anthropic/Claude%20Code/deferred-tools.md) · [Docs assistant](Anthropic/Claude%20Code/claude-code-docs-assistant.md) · [Bundled skills](Anthropic/Claude%20Code/bundled-skills/) | | Published (`claude_behavior` at release date, not updated) | [Opus 4.8](Anthropic/Official/2026-05-28-claude-opus-4.8.md) · [Opus 4.7](Anthropic/Official/2026-04-16-claude-opus-4.7.md) · [Opus 4.6](Anthropic/Official/2026-02-05-claude-opus-4.6.md) · [Sonnet 4.6](Anthropic/Official/2026-02-17-claude-sonnet-4.6.md) · [All versions](Anthropic/Official/) | | Without tools | [Opus 4.6](Anthropic/claude-opus-4.6-no-tools.md) · [Sonnet 4.6](Anthropic/claude-sonnet-4.6-no-tools.md) | | Raw prompts | [Opus 4.6](Anthropic/raw/claude-opus-4.6-raw.md) · [Opus 4.6 (no tools)](Anthropic/raw/claude-opus-4.6-no-tools-raw.md) · [Sonnet 4.6](Anthropic/raw/claude-sonnet-4.6-raw.md) · [Sonnet 4.6 (no tools)](Anthropic/raw/claude-sonnet-4.6-no-tools-raw.md) | | Visualize | [Visualization](Anthropic/visualize.md) | | Opus 4.5 | [System prompt](Anthropic/old/claude-opus-4.5.md) | | Sonnet 4.5 | [System prompt](Anthropic/old/claude-4.5-sonnet.md) | | Sonnet 4 | [System prompt](Anthropic/old/claude-sonnet-4.md) | | Opus 4.1 Thinking | [System prompt](Anthropic/old/claude-4.1-opus-thinking.md) | | Sonnet 3.7 | [System prompt](Anthropic/old/claude-3.7-sonnet.md) · [With tools](Anthropic/old/claude-3.7-sonnet-w-tools.md) · [Full w/ tools](Anthropic/old/claude-3.7-full-system-message-with-all-tools.md) · [Human-readable](Anthropic/old/claude-3.7-sonnet-full-system-message-humanreadable.md) | </details>  ## OpenAI — ChatGPT | Model | Prompt | |-------|--------| | **GPT-5.5** | [**Thinking**](OpenAI/gpt-5.5-thinking.md) · [**Instant**](OpenAI/gpt-5.5-instant.md) · [API](OpenAI/gpt-5.5-api.md) · [Pro API](OpenAI/gpt-5.5-pro-api.md) · [**Codex**](OpenAI/Codex/gpt-5.5.md) · [Friendly](OpenAI/Codex/personality_friendly_gpt-5.5.md) · [Pragmatic](OpenAI/Codex/personality_pragmatic_gpt-5.5.md) | | **GPT-5.4** | [**API**](OpenAI/gpt-5.4-api.md) · [**Thinking**](OpenAI/gpt-5.4-thinking.md) · [**Codex**](OpenAI/Codex/gpt-5.4.md) · [Codex Mini](OpenAI/Codex/gpt-5.4-mini.md) | | **GPT-5.3** | [**Codex**](OpenAI/Codex/gpt-5.3-codex.md) · [Spark](OpenAI/Codex/gpt-5.3-codex-spark.md) · [Codex API](OpenAI/gpt-5.3-codex-api.md) · [Chat API](OpenAI/gpt-5.3-chat-api.md) · [Instant](OpenAI/gpt-5.3-instant.md) | | **Codex CLI** | [Per-model prompts](OpenAI/Codex/) · [Spark](OpenAI/Codex/gpt-5.3-codex-spark.md) · [Plan mode](OpenAI/Codex/plan_mode.md) · [Personas](OpenAI/Codex/personality_friendly.md) · [Auto-review](OpenAI/Codex/codex-auto-review.md) | | **Tools** | [Web search](OpenAI/tool-web-search.md) · [Deep research](OpenAI/tool-deep-research.md) · [Python](OpenAI/tool-python.md) · [Python code](OpenAI/tool-python-code.md) · [Canvas](OpenAI/tool-canvas-canmore.md) · [Image gen](OpenAI/tool-create-image-image_gen.md) · [Memory](OpenAI/tool-memory-bio.md) · [Advanced memory](OpenAI/tool-advanced-memory.md) · [File search](OpenAI/tool-file_search.md) | | **Policies** | [Image safety](OpenAI/prompt-image-safety-policies.md) · [Automation context](OpenAI/prompt-automation-context.md) | <details><summary>Older models & variants</summary> | | | |--|--| | GPT-5.2 | [Mini (free)](OpenAI/gpt-5.2-mini-free-account.md) · [Thinking](OpenAI/gpt-5.2-thinking.md) · [Codex](OpenAI/Codex/gpt-5.2-codex.md) | | o4-mini | [System prompt](OpenAI/o4-mini.md) · [High](OpenAI/o4-mini-high.md) | | o3 | [System prompt](OpenAI/o3.md) | | ChatGPT Atlas | [System prompt](OpenAI/chatgpt-atlas.md) | | GPT-5.1 personalities | [Default](OpenAI/gpt-5.1-default.md) · [Friendly](OpenAI/gpt-5.1-friendly.md) · [Professional](OpenAI/gpt-5.1-professional.md) · [Candid](OpenAI/gpt-5.1-candid.md) · [Cynical](OpenAI/gpt-5.1-cynical.md) · [Efficient](OpenAI/gpt-5.1-efficient.md) · [Nerdy](OpenAI/gpt-5.1-nerdy.md) · [Quirky](OpenAI/gpt-5.1-quirky.md) | | GPT-5 | [Agent mode](OpenAI/chatgpt-gpt-5-agent-mode.md) · [Thinking](OpenAI/gpt-5-thinking.md) · [Cynic](OpenAI/gpt-5-cynic-personality.md) · [Listener](OpenAI/gpt-5-listener-personality.md) · [Nerdy](OpenAI/gpt-5-nerdy-personality.md) · [Robot](OpenAI/gpt-5-robot-personality.md) · [Codex](OpenAI/Codex/gpt-5-codex.md) · [Codex Mini](OpenAI/Codex/gpt-5-codex-mini.md) | | GPT-4.5 | [System prompt](OpenAI/gpt-4.5.md) | | GPT-4.1 | [Full](OpenAI/gpt-4.1.md) · [Mini](OpenAI/gpt-4.1-mini.md) | | GPT-4o | [System prompt](OpenAI/gpt-4o.md) · [WhatsApp](OpenAI/gpt-4o-whatsapp.md) · [Advanced voice](OpenAI/gpt-4o-advanced-voice-mode.md) · [Legacy voice](OpenAI/gpt-4o-legacy-voice-mode.md) | | Monday GPT | [System prompt](OpenAI/monday-gpt.md) | | GPT-4o new personality | [System prompt](OpenAI/4o-2025-09-03-new-personality.md) | | Study and learn | [System prompt](OpenAI/study-and-learn.md) | | Image safety policies | [System prompt](OpenAI/image-safety-policies.md) | | API variants | [GPT-5 reasoning (high)](OpenAI/API/gpt-5-reasoning-effort-high-api.md) · [o3 high](OpenAI/API/o3-high-api.md) · [o3 med](OpenAI/API/o3-medium-api.md) · [o3 low](OpenAI/API/o3-low-api.md) · [o4-mini high](OpenAI/API/o4-mini-high.md) · [o4-mini med](OpenAI/API/o4-mini-medium-api.md) · [o4-mini low](OpenAI/API/o4-mini-low-api.md) | | Old o4-mini | [System prompt](OpenAI/Old/chatgpt.com-o4-mini.md) | | Codex (older) | [GPT-5](OpenAI/Codex/gpt-5.md) · [GPT-5.1](OpenAI/Codex/gpt-5.1.md) · [GPT-5.1 Codex](OpenAI/Codex/gpt-5.1-codex.md) · [GPT-5.1 Mini](OpenAI/Codex/gpt-5.1-codex-mini.md) · [GPT-5.1 Max](OpenAI/Codex/gpt-5.1-codex-max.md) · [GPT-5.2](OpenAI/Code
Lo que la gente pregunta sobre system_prompts_leaks
¿Qué es asgeirtj/system_prompts_leaks?
+
asgeirtj/system_prompts_leaks es tools para el ecosistema de Claude AI. Extracted system prompts from Anthropic - Claude Fable 5, Opus 4.8, Claude Code, Claude Design. OpenAI - ChatGPT 5.5 Thinking, GPT 5.5 Instant, Codex. Google - Gemini 3.5 Flash, 3.1 Pro, Antigravity. xAI - Grok, Cursor, Copilot, VS Code, Perplexity, and more. Updated regularly. Tiene 41.9k estrellas en GitHub y se actualizó por última vez today.
¿Cómo se instala system_prompts_leaks?
+
Puedes instalar system_prompts_leaks clonando el repositorio (https://github.com/asgeirtj/system_prompts_leaks) o siguiendo las instrucciones del README en GitHub. ClaudeWave también te ofrece bloques de instalación rápida en esta misma página.
¿Es seguro usar asgeirtj/system_prompts_leaks?
+
Nuestro agente de seguridad ha analizado asgeirtj/system_prompts_leaks y le ha asignado un Trust Score de 73/100 (tier: OK). Revisa el desglose completo de comprobaciones superadas y flags en esta página.
¿Quién mantiene asgeirtj/system_prompts_leaks?
+
asgeirtj/system_prompts_leaks es mantenido por asgeirtj. La última actividad registrada en GitHub es de today, con 26 issues abiertos.
¿Hay alternativas a system_prompts_leaks?
+
Sí. En ClaudeWave puedes explorar tools similares en /categories/tools, ordenados por popularidad o actividad reciente.
Despliega system_prompts_leaks en tu cloud
Lleva este repo a producción en minutos. Cada plataforma genera su propio entorno con variables de entorno editables.
¿Mantienes este repo? Añade un badge a tu README
Pega el badge en tu README de GitHub para mostrar que está auditado por ClaudeWave. Cada badge enlaza de vuelta a esta página y muestra el Trust Score actual.
[](https://claudewave.com/repo/asgeirtj-system-prompts-leaks)<a href="https://claudewave.com/repo/asgeirtj-system-prompts-leaks"><img src="https://claudewave.com/api/badge/asgeirtj-system-prompts-leaks" alt="Featured on ClaudeWave: asgeirtj/system_prompts_leaks" width="320" height="64" /></a>Más Tools
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
An AI SKILL that provide design intelligence for building professional UI/UX multiple platforms
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.
A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies