{"id":31727346,"url":"https://github.com/keyboardcowboy/ai-context-kitchen","last_synced_at":"2026-02-14T21:02:42.830Z","repository":{"id":314326849,"uuid":"1055005124","full_name":"KeyboardCowboy/ai-context-kitchen","owner":"KeyboardCowboy","description":"A system for organizing AI context window rules and files to create efficient and accurate agents.","archived":false,"fork":false,"pushed_at":"2025-09-12T02:20:23.000Z","size":68,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-09T06:22:13.721Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/KeyboardCowboy.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"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":"2025-09-11T16:20:57.000Z","updated_at":"2025-09-15T12:37:10.000Z","dependencies_parsed_at":"2025-09-11T22:09:15.199Z","dependency_job_id":"f02ffc23-2339-414c-a199-9541bbe1103e","html_url":"https://github.com/KeyboardCowboy/ai-context-kitchen","commit_stats":null,"previous_names":["keyboardcowboy/ai-context-kitchen"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/KeyboardCowboy/ai-context-kitchen","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KeyboardCowboy%2Fai-context-kitchen","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KeyboardCowboy%2Fai-context-kitchen/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KeyboardCowboy%2Fai-context-kitchen/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KeyboardCowboy%2Fai-context-kitchen/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/KeyboardCowboy","download_url":"https://codeload.github.com/KeyboardCowboy/ai-context-kitchen/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/KeyboardCowboy%2Fai-context-kitchen/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29455601,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-14T15:52:44.973Z","status":"ssl_error","status_checked_at":"2026-02-14T15:52:11.208Z","response_time":53,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":[],"created_at":"2025-10-09T06:20:53.900Z","updated_at":"2026-02-14T21:02:42.809Z","avatar_url":"https://github.com/KeyboardCowboy.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# AI Agent Context Framework\n\n![Version](https://img.shields.io/badge/version-1.0.0-blue.svg)\n![License](https://img.shields.io/badge/license-MIT-green.svg)\n![Framework](https://img.shields.io/badge/framework-Anthropic%20Claude-purple.svg)\n![Status](https://img.shields.io/badge/status-active-brightgreen.svg)\n\nA four-tiered system for maximizing AI agent accuracy and efficiency through structured context window seeding, built on Anthropic's prompt engineering best practices.\n\n## What This Framework Does\n\nInstead of writing one massive prompt, you build AI agents by combining four types of reusable components:\n\n1. **Role** - Who the agent is and what they know (direct prompts)\n2. **Rules** - How to accomplish specific tasks (XML structured)  \n3. **Background** - Essential facts and context (XML structured)\n4. **Interaction** - How the agent communicates (direct prompts)\n\n**Key Insight**: Separate expertise (WHAT) from communication style (HOW). This lets you mix and match - same expert knowledge with different personalities.\n\n## Why This Works\n\nBuilt on Anthropic's proven techniques:\n- **Direct prompts** for simple role definitions and communication\n- **XML structure** for complex, organized content\n- **Role prompting** for domain expertise\n- **Optimized** for Claude's capabilities\n\n## Project Structure\n\n```\ncontext-kitchen/\n├── 1-roles/           # Who the agent is (direct prompts)\n├── 2-background/      # Essential facts (XML structured)  \n├── 3-rules/           # How to do tasks (XML structured)\n└── 4-interaction/     # How to communicate (direct prompts)\n```\n\nEach folder contains:\n- **Root files**: No prefix (e.g., `seo-expert.md`)\n- **Subfolder files**: Use folder name as prefix (e.g., `example-seo-expert.md`)\n\n## Quick Start\n\n```bash\n# 1. Choose your components\nRole: SEO Expert\nRules: research-analysis, content-optimization  \nBackground: business-context\nInteraction: collaborative-learning\n\n# 2. Combine in your prompt\nYou are a [Role] with expertise in [Background]. \nFollow these [Rules] and communicate using [Interaction] style.\n```\n\n## Installation\n\nThis framework works with any AI system that supports:\n- Direct system prompts\n- XML structured content  \n- Multi-component context loading\n\n### Prerequisites\n- Basic understanding of prompt engineering\n- Text editor for managing components\n\n## How to Use\n\n1. **Pick a Role** - Choose one that matches your objective\n2. **Add Rules** - Select 2-4 that support the role's tasks  \n3. **Include Background** - Add relevant facts and context\n4. **Set Interaction Style** - Choose how the agent communicates\n\n## Component Guidelines\n\n### Roles (Choose 1)\n- **What**: Domain expertise and professional identity\n- **Focus**: WHO and WHAT, not HOW (that's for rules)\n- **Format**: Direct prompts, no XML\n\n### Rules (Choose 2-4)  \n- **What**: Complete, standalone processes\n- **Focus**: HOW to accomplish specific tasks\n- **Format**: XML structured for complex procedures\n\n### Background (Choose relevant ones)\n- **What**: Essential facts and context\n- **Focus**: Lean, fact-focused content without fluff\n- **Format**: XML structured for organized knowledge\n\n### Interaction (Choose 1)\n- **What**: Communication style and personality  \n- **Focus**: HOW the agent behaves, not WHAT it knows\n- **Format**: Direct prompts, no XML\n\n## Example Configurations\n\n**Research \u0026 Analysis**\n- Role: SEO Expert + Research rules + Business background + Interview mode\n\n**Content Creation**  \n- Role: Content Specialist + Writing rules + Brand background + Quick artifact\n\n**Financial Planning**\n- Role: Financial Advisor + Planning rules + Family background + Plan-act mode\n\n## Key Principles\n\n**Separation of Concerns**\n- **Roles**: WHO and WHAT (expertise)\n- **Rules**: HOW (processes)  \n- **Background**: Facts and context\n- **Interaction**: HOW (communication)\n\n**Conflict Resolution**\n- Role takes precedence for decisions\n- Background facts override assumptions\n- Rules override general guidance\n- Interaction style overrides communication\n\n## Quality Checklist\n\n- [ ] One role selected\n- [ ] 2-4 relevant rules included  \n- [ ] Appropriate background knowledge added\n- [ ] Compatible interaction style chosen\n- [ ] No major conflicts between tiers\n- [ ] Complete coverage for intended purpose\n\n## Benefits\n\n- **Modular**: Mix and match components\n- **Reusable**: Build a library of configurations\n- **Consistent**: Standardized format\n- **Efficient**: Optimized context window usage\n- **AI-Friendly**: Built for maximum performance\n\n## Contributing\n\nWe welcome contributions! Here's how to get started:\n\n1. **Fork the repository**\n2. **Create a new component** following our naming conventions\n3. **Test your component** with different configurations\n4. **Submit a pull request** with clear descriptions\n\n### Component Guidelines\n- **Roles**: Focus on WHO and WHAT, not HOW\n- **Rules**: Complete, standalone processes\n- **Background**: Lean, fact-focused content\n- **Interaction**: Communication style only\n\n### Naming Conventions\n- **Root files**: No prefix (e.g., `seo-expert.md`)\n- **Subfolder files**: Use folder name as prefix (e.g., `example-seo-expert.md`)\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n\n## Support\n\n- 📖 [Documentation](docs/)\n- 💬 [Discussions](https://github.com/your-org/context-kitchen/discussions)\n- 🐛 [Issues](https://github.com/your-org/context-kitchen/issues)\n- 📧 [Email](mailto:support@example.com)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeyboardcowboy%2Fai-context-kitchen","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkeyboardcowboy%2Fai-context-kitchen","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeyboardcowboy%2Fai-context-kitchen/lists"}