{"id":30762698,"url":"https://github.com/tobrun/agent-factory","last_synced_at":"2026-02-10T21:32:08.675Z","repository":{"id":312146413,"uuid":"1046504317","full_name":"tobrun/agent-factory","owner":"tobrun","description":" Claude Code sub-agent that specializes in creating new specialist sub-agents from brief descriptions. This factory generates complete, ready-to-use sub-agent configuration files with proper scoping, documentation, and research workflows.","archived":false,"fork":false,"pushed_at":"2025-08-28T19:53:12.000Z","size":21,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-08-29T01:37:26.923Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"HTML","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/tobrun.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}},"created_at":"2025-08-28T19:33:33.000Z","updated_at":"2025-08-28T19:53:16.000Z","dependencies_parsed_at":"2025-08-29T01:37:29.133Z","dependency_job_id":"72fa3749-4bdf-4e20-8faa-7e77a7a0ca0c","html_url":"https://github.com/tobrun/agent-factory","commit_stats":null,"previous_names":["tobrun/agent-factory"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/tobrun/agent-factory","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tobrun%2Fagent-factory","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tobrun%2Fagent-factory/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tobrun%2Fagent-factory/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tobrun%2Fagent-factory/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/tobrun","download_url":"https://codeload.github.com/tobrun/agent-factory/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/tobrun%2Fagent-factory/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29317956,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-10T20:44:44.282Z","status":"ssl_error","status_checked_at":"2026-02-10T20:44:43.393Z","response_time":65,"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":"2025-09-04T15:10:12.356Z","updated_at":"2026-02-10T21:32:08.664Z","avatar_url":"https://github.com/tobrun.png","language":"HTML","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Agent Factory\n\nA Claude Code sub-agent that specializes in creating new specialist sub-agents from brief descriptions. This factory generates complete, ready-to-use sub-agent configuration files with proper scoping, documentation, and research workflows.\n\n## Overview\n\nThe Agent Factory is a **sub-agent architect \u0026 planner** that transforms user requirements into research-focused specialist sub-agents. It follows a structured framework to ensure each generated agent is properly scoped, well-documented, and optimized for token efficiency.\n\n## Key Features\n\n- **Automated Agent Generation**: Creates complete sub-agent configs from brief descriptions\n- **Research-Only Focus**: All generated agents are scoped as researchers and planners only\n- **Context Management**: Implements persistent documentation patterns to reduce token usage\n- **Framework Compliance**: Ensures all agents follow Claude Code sub-agent standards\n- **Documentation Integration**: Automatically creates supporting research artifacts\n\n## How It Works\n\n### Input\nProvide a brief description of the specialist expertise needed:\n```\n\"Use agent factory to create a Firebase expert for authentication and Firestore integration\"\n\"Create a React Native specialist for mobile development using agent factory\"\n\"Build a agent factory agent for Docker deployment\"\n```\n\n### Output\nThe factory generates:\n1. **Sub-agent config file**: `.claude/agents/\u003cgenerated-agent-name\u003e.md`\n2. **Implementation plan**: `.claude/docs/\u003cservice\u003e-\u003ctopic\u003e-plan-\u003ctimestamp\u003e.md`\n3. **Context session**: `.claude/docs/context-session-\u003cdate\u003e.md`\n\n## Generated Agent Structure\n\nEach generated agent includes:\n\n### Frontmatter Configuration\n- **Name**: Kebab-case naming (e.g., `firebase-expert`, `react-native-specialist`)\n- **Description**: Action-oriented delegation guidance\n- **Tools**: Minimal research-focused toolset\n- **Model**: Optimized model selection\n\n### System Prompt Sections\n- **Research-Only Boundaries**: Clear scope limitations\n- **Context Management Pattern**: File structure and workflows\n- **Knowledge Areas**: Service-specific expertise domains\n- **Procedure Checklist**: Step-by-step research methodology\n- **Output Contract**: Standardized deliverable formats\n\n## Agent Scoping Framework\n\nThe factory applies these design principles:\n\n### ✅ Generated Agents Will:\n- Research documentation \u0026 best practices\n- Analyze repository integration points (read-only)\n- Produce detailed implementation plans\n- Create persistent documentation in `.claude/docs/`\n- Optimize for token efficiency through context reuse\n\n### ❌ Generated Agents Will NOT:\n- Implement or execute code\n- Modify runtime source files\n- Make direct system changes\n- Override the research-only boundary\n\n## File Structure\n\nGenerated agents maintain this documentation pattern:\n\n```\n.claude/\n├── agents/\n│   └── \u003cgenerated-agent-name\u003e.md\n└── docs/\n    ├── context-session-[timestamp].md\n    ├── [service]-[topic]-plan-[timestamp].md\n    └── task/\n        └── [specific-research-files].md\n```\n\n## Usage Examples\n\n### Creating a Database Expert\n```\nRequest: \"Create a PostgreSQL expert for query optimization and schema design\"\n\nOutput:\n- .claude/agents/postgresql-expert.md\n- .claude/docs/postgresql-optimization-plan-20250828-1430.md\n- Updated context session with research findings\n```\n\n### Building an API Specialist\n```\nRequest: \"Need a REST API expert for authentication and rate limiting\"\n\nOutput:\n- .claude/agents/api-security-expert.md\n- .claude/docs/api-auth-plan-20250828-1445.md\n- Comprehensive research on auth patterns and rate limiting\n```\n\n## Quality Assurance\n\nThe factory includes built-in validation:\n\n### Agent Prioritization Matrix\n- **High Priority**: Complex, fast-changing, or central technologies\n- **Medium Priority**: Stable technologies with specific integration needs\n- **Low Priority**: Simple or well-documented technologies\n\n### Implementation Checklist\n- [ ] Define service scope\n- [ ] List documentation sources\n- [ ] Identify MCP tools/APIs\n- [ ] Specify knowledge areas\n- [ ] Set context workflow \u0026 files\n- [ ] Enforce research-only boundary\n- [ ] Test with sample research task\n- [ ] Integrate into parent workflow\n\n## Best Practices\n\n### Token Optimization\n- Reuses existing context from `.claude/docs/`\n- Cites documentation versions to avoid re-fetching\n- Maintains session history for incremental research\n\n### Documentation Standards\n- Consistent formatting across all generated plans\n- Explicit risk and assumption documentation\n- Implementation-ready guidance without actual code\n- Proper linking and versioning of source materials\n\n## Getting Started\n\n1. **Invoke the Factory**: Describe the specialist expertise needed\n2. **Review the Plan**: Start with the generated plan file in `.claude/docs/`\n3. **Use the Agent**: Delegate tasks to the newly created specialist\n4. **Iterate**: Refine based on usage patterns and feedback\n\n## Technical Details\n\n### Available Tools\n- **Core**: Read, Write, Glob, Grep\n- **Documentation**: WebFetch, context7 library tools\n- **Multi-file Operations**: MultiEdit for complex configurations\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftobrun%2Fagent-factory","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftobrun%2Fagent-factory","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftobrun%2Fagent-factory/lists"}