{"id":47917280,"url":"https://github.com/anasfik/claude-code-prompts-reference","last_synced_at":"2026-04-08T09:00:45.274Z","repository":{"id":348584425,"uuid":"1198806625","full_name":"anasfik/claude-code-prompts-reference","owner":"anasfik","description":"Leaked Claude code research material  (how it does everything)","archived":false,"fork":false,"pushed_at":"2026-04-01T20:35:18.000Z","size":69,"stargazers_count":5,"open_issues_count":0,"forks_count":2,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-04-07T08:03:32.850Z","etag":null,"topics":["claude","claude-code","code","data","leak","leaks","prompts","research","skills","system-prompts","tools"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/anasfik.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":"audit/prompt-source-audit.md","citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-01T19:23:48.000Z","updated_at":"2026-04-05T11:26:13.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/anasfik/claude-code-prompts-reference","commit_stats":null,"previous_names":["anasfik/claude-code-prompts-reference"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/anasfik/claude-code-prompts-reference","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anasfik%2Fclaude-code-prompts-reference","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anasfik%2Fclaude-code-prompts-reference/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anasfik%2Fclaude-code-prompts-reference/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anasfik%2Fclaude-code-prompts-reference/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/anasfik","download_url":"https://codeload.github.com/anasfik/claude-code-prompts-reference/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/anasfik%2Fclaude-code-prompts-reference/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31547845,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-07T16:28:08.000Z","status":"online","status_checked_at":"2026-04-08T02:00:06.127Z","response_time":54,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["claude","claude-code","code","data","leak","leaks","prompts","research","skills","system-prompts","tools"],"created_at":"2026-04-04T05:45:33.736Z","updated_at":"2026-04-08T09:00:45.003Z","avatar_url":"https://github.com/anasfik.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Claude Code — Complete AI Prompt Reference\n\nCurated reference of prompt text used in Claude Code, extracted from source and organized into standalone markdown docs.\n\n---\n\n## What this repo contains\n\nThis repository documents prompt logic across Claude Code across all major prompt surfaces:\n\n- **System prompt** assembly and policy sections\n- **Tool prompts** (`DESCRIPTION` / `PROMPT` and prompt builders)\n- **Skill prompts** (bundled slash-command workflows)\n- **Browser automation** prompt guidance\n- **Command-specific prompts** (`/review`, `/ultraplan`)\n- **Additional command prompts** (`/init`, `/insights`)\n- **Memory system prompts** (selection, extraction, session memory, consolidation)\n- **Service-level prompts** (compact, summaries, suggestions, search/title)\n- **Internal agent prompts** (verification/statusline/teammate/agent generation)\n- **Permission/classifier prompts** (permission explainer, auto-mode classifier surfaces)\n- **Interface prompts** (output styles, teleport title/branch)\n\n---\n\n## Quick navigation\n\n- [auto-mode.md](commands/auto-mode.md)\n- [security-instruction.md](system-prompt/security-instruction.md)\n\n### Tools\n\n### Coverage audit\n\n- [prompt-source-audit.md](audit/prompt-source-audit.md)\n- [file-tools.md](tools/file-tools.md)\n- [shell-tools.md](tools/shell-tools.md)\n- [search-tools.md](tools/search-tools.md)\n- [agent-coordination.md](tools/agent-coordination.md)\n- [planning-tools.md](tools/planning-tools.md)\n- [task-management.md](tools/task-management.md)\n- [team-coordination.md](tools/team-coordination.md)\n- [scheduling.md](tools/scheduling.md)\n- [worktree-tools.md](tools/worktree-tools.md)\n- [mcp-tools.md](tools/mcp-tools.md)\n- [misc-tools.md](tools/misc-tools.md)\n\n### Skills\n\n- [batch.md](skills/batch.md)\n- [simplify.md](skills/simplify.md)\n- [debug.md](skills/debug.md)\n- [stuck.md](skills/stuck.md)\n- [remember.md](skills/remember.md)\n- [skillify.md](skills/skillify.md)\n- [loop.md](skills/loop.md)\n- [schedule.md](skills/schedule.md)\n- [claude-api.md](skills/claude-api.md)\n- [update-config.md](skills/update-config.md)\n\n### Command prompt surfaces\n\n- [review.md](commands/review.md)\n- [ultraplan.md](commands/ultraplan.md)\n- [init-and-insights.md](commands/init-and-insights.md)\n\n### Memory prompts\n\n- [memory-prompts.md](memory/memory-prompts.md)\n\n### Service prompts\n\n- [session-and-meta-prompts.md](services/session-and-meta-prompts.md)\n\n### Internal agent prompts\n\n│ └── auto-mode.md\n\n- [internal-agent-prompts.md](internal-agents/internal-agent-prompts.md)\n\n### Permission/classifier prompts\n\n- [permission-and-classifier-prompts.md](permissions/permission-and-classifier-prompts.md)\n\n### Interface prompts\n\n- [output-and-teleport-prompts.md](interface/output-and-teleport-prompts.md)\n\n# Claude Code — Complete AI Prompt Reference\n\nCurated reference of prompt text used in Claude Code, extracted from source and organized into standalone markdown docs.\n\n---\n\n## What this repo contains\n\nThis repository documents prompt logic across Claude Code across all major prompt surfaces:\n\n- **System prompt** assembly and policy sections\n- **Tool prompts** (`DESCRIPTION` / `PROMPT` and prompt builders)\n- **Skill prompts** (bundled slash-command workflows)\n- **Browser automation** prompt guidance\n- **Command-specific prompts** (`/review`, `/ultraplan`, `/init`, `/insights`, `auto-mode`)\n- **Memory system prompts** (selection, extraction, session memory, consolidation)\n- **Service-level prompts** (compact, summaries, suggestions, search/title)\n- **Internal agent prompts** (verification/statusline/teammate/agent generation)\n- **Permission/classifier prompts** (permission explainer, auto-mode classifier surfaces)\n- **Interface prompts** (output styles, teleport title/branch)\n- **Source coverage audit** for prompt-bearing files\n\n---\n\n## Quick navigation\n\n### System prompt\n\n- [main-system-prompt.md](system-prompt/main-system-prompt.md)\n- [security-instruction.md](system-prompt/security-instruction.md)\n- [effective-system-prompt-selection.md](system-prompt/effective-system-prompt-selection.md)\n\n### Tools\n\n- [file-tools.md](tools/file-tools.md)\n- [shell-tools.md](tools/shell-tools.md)\n- [search-tools.md](tools/search-tools.md)\n- [agent-coordination.md](tools/agent-coordination.md)\n- [planning-tools.md](tools/planning-tools.md)\n- [task-management.md](tools/task-management.md)\n- [team-coordination.md](tools/team-coordination.md)\n- [scheduling.md](tools/scheduling.md)\n- [worktree-tools.md](tools/worktree-tools.md)\n- [mcp-tools.md](tools/mcp-tools.md)\n- [misc-tools.md](tools/misc-tools.md)\n\n### Skills\n\n- [batch.md](skills/batch.md)\n- [simplify.md](skills/simplify.md)\n- [debug.md](skills/debug.md)\n- [stuck.md](skills/stuck.md)\n- [remember.md](skills/remember.md)\n- [skillify.md](skills/skillify.md)\n- [loop.md](skills/loop.md)\n- [schedule.md](skills/schedule.md)\n- [claude-api.md](skills/claude-api.md)\n- [update-config.md](skills/update-config.md)\n\n### Command prompt surfaces\n\n- [review.md](commands/review.md)\n- [ultraplan.md](commands/ultraplan.md)\n- [init-and-insights.md](commands/init-and-insights.md)\n- [auto-mode.md](commands/auto-mode.md)\n\n### Memory prompts\n\n- [memory-prompts.md](memory/memory-prompts.md)\n\n### Service prompts\n\n- [session-and-meta-prompts.md](services/session-and-meta-prompts.md)\n\n### Internal agent prompts\n\n- [internal-agent-prompts.md](internal-agents/internal-agent-prompts.md)\n\n### Permission/classifier prompts\n\n- [permission-and-classifier-prompts.md](permissions/permission-and-classifier-prompts.md)\n\n### Interface prompts\n\n- [output-and-teleport-prompts.md](interface/output-and-teleport-prompts.md)\n\n### Other prompt surfaces\n\n- [chrome-prompts.md](browser-automation/chrome-prompts.md)\n- [buddy-prompt.md](companion/buddy-prompt.md)\n\n### Coverage audit\n\n- [prompt-source-audit.md](audit/prompt-source-audit.md)\n\n---\n\n## Repository layout\n\n```\nclaude-code-system-promtps/\n├── README.md\n├── .gitignore\n│\n├── system-prompt/\n│   ├── main-system-prompt.md\n│   ├── security-instruction.md\n│   └── effective-system-prompt-selection.md\n│\n├── tools/\n│   ├── file-tools.md\n│   ├── shell-tools.md\n│   ├── search-tools.md\n│   ├── agent-coordination.md\n│   ├── planning-tools.md\n│   ├── task-management.md\n│   ├── team-coordination.md\n│   ├── scheduling.md\n│   ├── worktree-tools.md\n│   ├── mcp-tools.md\n│   └── misc-tools.md\n│\n├── skills/\n│   ├── batch.md\n│   ├── simplify.md\n│   ├── debug.md\n│   ├── stuck.md\n│   ├── remember.md\n│   ├── skillify.md\n│   ├── loop.md\n│   ├── schedule.md\n│   ├── claude-api.md\n│   └── update-config.md\n│\n├── browser-automation/\n│   └── chrome-prompts.md\n│\n├── companion/\n│   └── buddy-prompt.md\n│\n├── commands/\n│   ├── review.md\n│   ├── ultraplan.md\n│   ├── init-and-insights.md\n│   └── auto-mode.md\n│\n├── memory/\n│   └── memory-prompts.md\n│\n├── services/\n│   └── session-and-meta-prompts.md\n│\n├── internal-agents/\n│   └── internal-agent-prompts.md\n│\n├── permissions/\n│   └── permission-and-classifier-prompts.md\n│\n├── interface/\n│   └── output-and-teleport-prompts.md\n│\n└── audit/\n    └── prompt-source-audit.md\n```\n\n---\n\n## Prompt architecture (high-level)\n\nClaude Code prompt behavior is split into three primary layers, plus supporting prompt surfaces:\n\n1. **Global system prompt** (`src/constants/prompts.ts`)\n2. **Tool prompts** (`src/tools/*/prompt.ts`)\n3. **Skill / command prompts** (`src/skills/bundled/*.ts`, `src/commands/*.ts`)\n\nSupporting prompt surfaces include memory extraction/selection prompts, compaction/suggestion prompts, classifier prompts, and internal agent prompts.\n\n---\n\n## Coverage summary\n\n- Core system prompt and security policy\n- Effective system-prompt selection logic (`buildEffectiveSystemPrompt`)\n- Full tool prompt set used in the main runtime\n- Bundled skills and command entry prompts (including `/update-config`)\n- Memory subsystem prompts (selection, extraction, session notes, consolidation)\n- Background/service prompts (compact, away-summary, tool-use-summary, suggestions, semantic search, title generation)\n- Internal agent prompts and teammate addenda\n- Permission/classifier prompt surfaces\n- UI/interface prompt surfaces (output styles, teleport)\n- Source-to-doc audit mapping for prompt-bearing files\n\n---\n\n## How to use this reference\n\n1. Start in **system-prompt/** for global behavior.\n2. Use **tools/** for callable tool instructions.\n3. Use **skills/** and **commands/** for workflow prompts.\n4. Use **memory/** and **services/** for background model prompt behavior.\n5. Use **internal-agents/**, **permissions/**, and **interface/** for internal/runtime-specific prompt surfaces.\n6. Use **audit/** to verify source-to-doc coverage.\n\n---\n\n## Scope note\n\n- This reference tracks prompt material from Claude Code source files.\n- No external repository is used as a source of truth for this folder.\n- If an external URL/name appears in quoted prompt text, it is preserved only because it exists in source.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanasfik%2Fclaude-code-prompts-reference","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fanasfik%2Fclaude-code-prompts-reference","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fanasfik%2Fclaude-code-prompts-reference/lists"}