{"id":31770984,"url":"https://github.com/pigmej/opencode_config","last_synced_at":"2025-10-10T03:18:28.584Z","repository":{"id":316957101,"uuid":"1058935554","full_name":"pigmej/opencode_config","owner":"pigmej","description":null,"archived":false,"fork":false,"pushed_at":"2025-10-04T23:00:01.000Z","size":65,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-05T01:05:19.714Z","etag":null,"topics":["agentic-ai","opencode"],"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/pigmej.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2025-09-17T18:49:01.000Z","updated_at":"2025-09-29T18:28:16.000Z","dependencies_parsed_at":"2025-09-27T21:21:46.962Z","dependency_job_id":"7b5c4be6-c526-43ff-8b1f-9c11680c3118","html_url":"https://github.com/pigmej/opencode_config","commit_stats":null,"previous_names":["pigmej/opencode_config"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/pigmej/opencode_config","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pigmej%2Fopencode_config","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pigmej%2Fopencode_config/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pigmej%2Fopencode_config/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pigmej%2Fopencode_config/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pigmej","download_url":"https://codeload.github.com/pigmej/opencode_config/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pigmej%2Fopencode_config/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279002583,"owners_count":26083420,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-10T02:00:06.843Z","response_time":62,"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":["agentic-ai","opencode"],"created_at":"2025-10-10T03:18:27.033Z","updated_at":"2025-10-10T03:18:28.579Z","avatar_url":"https://github.com/pigmej.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# OpenCode Agent Orchestration Framework\n\nA structured framework for developing software features through coordinated agent collaboration, architectural analysis, and systematic implementation.\n\n## Overview\n\nOpenCode provides two workflow types for different development needs:\n\n### Simple Workflow (Tasks)\n1. **Task Creation** (`/auto_task`) - Create structured task files from discussions\n2. **Planning** (`/auto_plan`) - Generate comprehensive development plans with architectural analysis\n3. **Orchestration** (`/orch`) - Execute the plan using coordinated agents\n\n### Feature Workflow (Complex Projects)\n1. **Feature Creation** (`/auto_feature`) - Create feature specification and architecture\n2. **Feature Decomposition** (`/feature_decompose`) - Break feature into implementable tasks\n3. **Task Planning** (`/auto_plan`) - Create detailed plans for each task\n4. **Orchestration** (`/orch`) - Execute each task plan\n\nSee [FEATURE_WORKFLOW.md](FEATURE_WORKFLOW.md) for detailed feature workflow documentation.\n\n## Stage 1: Task Creation (`/auto_task`)\n\n### Purpose\nCreates a structured task file based on discussion context, transforming conversation into actionable development requirements.\n\n### Usage\n```\n/auto_task [additional suggestions and ideas]\n```\n\n### Process\n1. **Task ID Generation**: Finds the highest numbered file in `./.task/` directory and adds 10\n2. **File Naming**: Creates `{new-task-id}-{short-task-title}.md` using kebab-case (max 7 words)\n3. **Content Structure**:\n   - User feedback from discussion\n   - Problem Statement\n   - Requirements\n   - Expected Outcome\n   - Additional suggestions (from arguments)\n   - Other important agreements (architectural decisions, direction, project context)\n\n### Output\n- Task file in `./.task/` directory\n- Clear file path confirmation\n\n### Example\nAfter discussing authentication improvements:\n```\n/auto_task \"Consider OAuth2 integration and multi-factor authentication\"\n```\nCreates: `./.task/0030-user-authentication-system.md`\n\n## Stage 2: Planning (`/auto_plan`)\n\n### Purpose\nGenerates comprehensive development plans with architectural analysis and implementation details.\n\n### Usage\n```\n/auto_plan ./.task/[filename].md @architect @sonnet\n```\n\n### Process Flow\n\n#### Phase 0: Setup \u0026 Context Extraction\n- **Purpose**: Validate input and extract context for all agents\n- **Activities**:\n  - Validate task file exists\n  - Extract filename components and set file paths\n  - Read and summarize task file (max 300 tokens)\n  - Extract feature architecture summary if exists (max 400 tokens)\n  - Prepare inline context for all subsequent phases\n- **Reliability**: Clear error handling and termination conditions\n\n#### Phase 1: Architectural Analysis\n- **Agent**: `architect`\n- **Retry Limit**: 3 attempts\n- **Input**: Inline context + task file reference\n- **Output**: `./.plan/arch_[basename].md`\n- **Activities**:\n  - Targeted web research (when needed)\n  - Task-specific architecture patterns\n  - Integration approach if dependencies exist\n- **Reliability**: Automatic retry with clear failure handling\n\n#### Phase 2: Implementation Planning\n- **Agent**: First specified agent (`@agent_1`)\n- **Retry Limit**: 3 attempts\n- **Input**: Inline context + architectural analysis\n- **Output**: `./.plan/[filename].md`\n- **Activities**:\n  - Component details and data structures\n  - API design following architectural patterns\n  - Integration strategy with prior tasks\n  - Testing strategy\n  - Development phases\n- **Reliability**: Automatic retry with clear failure handling\n\n#### Phase 3: Review\n- **Agent**: Second specified agent (`@agent_2`)\n- **Scoring System**: \n  - Implementation Feasibility (40%)\n  - Architectural Alignment (30%)\n  - Completeness (20%)\n  - Integration Quality (10%)\n- **Decision Logic**: 90%+ = PASS, \u003c90% = NEEDS REFINEMENT\n- **Activities**:\n  - Read architectural analysis and implementation plan\n  - Provide detailed percentage-based feedback\n  - Generate specific improvement recommendations\n- **Output**: Overall score + detailed breakdown + feedback\n\n#### Phase 4: Refinement Loop\n- **Trigger**: Overall score \u003c 90%\n- **Maximum Iterations**: 3 total\n- **Process**: \n  - Categorize issues (architectural vs implementation)\n  - Spawn appropriate agents (@agent_2 for architectural, @agent_1 for implementation)\n  - Use actual review feedback (not placeholders)\n  - Re-evaluate until 90%+ achieved or max iterations reached\n- **Reliability**: Clear iteration limits and graceful completion\n\n#### Phase 5: Final Completion\n- **Trigger**: 90%+ score OR max refinement iterations reached\n- **Output**: Comprehensive summary with quality metrics\n- **Quality Metrics**: \n  - Final overall score with detailed breakdown\n  - Iteration counts and improvement tracking\n  - File paths and architectural decisions summary\n\n### Output Files\n- `./.plan/arch_[filename].md` - Architectural analysis\n- `./.plan/[filename].md` - Implementation plan\n\n### Reliability Features 🛡️\n- **Retry Limits**: 3 attempts per phase with clear failure handling\n- **Linear Flow**: Predictable Phase 0→1→2→3→4→5 progression\n- **No Infinite Loops**: Eliminated \"repeat till file exists\" patterns\n- **Proper Variable Substitution**: Real feedback content instead of placeholders\n- **Graceful Degradation**: Best effort completion when limits reached\n\n### Example\n```\n/auto_plan ./.task/110-jwt-token-service.md @architect @sonnet\n```\n\n## Stage 3: Orchestration (`/orch`)\n\n### Purpose\nExecutes the development plan using coordinated agents with quality validation.\n\n### Usage\n```\n/orch ./.task/[filename].md ./.plan/[filename].md grok sonnet\n```\n\n### Process Flow\n\n#### Phase 0: Setup \u0026 Context Extraction\n- **Purpose**: Validate input and extract context for all agents\n- **Activities**:\n  - Read and summarize task file (max 300 tokens)\n  - Read and summarize plan file (max 400 tokens)\n  - Prepare inline context for all phases\n\n#### Phase 1: Plan Execution\n- **Agent**: First specified agent (`grok`)\n- **Input**: Inline context (task + plan summary)\n- **Activities**:\n  - Implement exactly as specified in plan summary\n  - Track modified files for incremental review\n  - No deviations or extra features\n  - No commits (code changes only)\n\n#### Phase 2: Implementation Review\n- **Agent**: Second specified agent (`sonnet`)\n- **Input**: Inline context + git diff of modified files only\n- **Evaluation Criteria**:\n  - Plan compliance (how many items implemented)\n  - Approach adherence (follows plan's methodology)\n  - Code quality and best practices\n  - Security and performance considerations\n  - Test coverage (if applicable)\n\n**Scoring System**:\n- 90-100%: Excellent execution, minor/no deviations\n- 70-89%: Good execution with some issues\n- 50-69%: Significant deviations or incomplete work\n- \u003c50%: Major failure to follow plan\n\n#### Phase 3: Iterative Improvement (if needed)\n- **Trigger**: Score \u003c 90%\n- **Process**: \n  1. Extract specific issues from review\n  2. First agent fixes only identified issues (with inline context)\n  3. Update modified files list\n  4. Second agent re-reviews (only changed files)\n  5. Repeat until 90%+ compliance\n\n#### Phase 4: Completion\n- **Output**: Final summary with compliance score and implementation details\n- **No commits**: User decides when to commit changes\n\n### Example\n```\n/orch ./.task/0030-user-authentication-system.md ./.plan/0030-user-authentication-system.md grok sonnet\n```\n\n## Complete Workflow Example\n\n### Simple Task Workflow\n```bash\n# 1. Create task from discussion\n/auto_task \"Add OAuth2 support and consider security best practices\"\n# Output: ./.task/0030-user-authentication-system.md\n\n# 2. Generate plan with architecture\n/auto_plan ./.task/0030-user-authentication-system.md @architect @sonnet\n# Output: \n# - ./.plan/arch_0030-user-authentication-system.md\n# - ./.plan/0030-user-authentication-system.md\n\n# 3. Execute plan\n/orch ./.task/0030-user-authentication-system.md ./.plan/0030-user-authentication-system.md grok sonnet\n# Output: Implemented code changes (not committed)\n```\n\n### Feature Workflow (Complex Projects)\n```bash\n# 1. Create feature with architecture\n/auto_feature Implement user authentication system with JWT and OAuth2\n\n# 2. Decompose into tasks\n/feature_decompose ./.feature/100-user-authentication-system.md\n\n# 3. Plan each task\n/auto_plan ./.task/100_1_10-jwt-service.md @architect @sonnet\n/auto_plan ./.task/100_1_20-oauth2-integration.md @architect @sonnet\n\n# 4. Execute each task\n/orch ./.task/100_1_10-jwt-service.md ./.plan/100_1_10-jwt-service.md grok sonnet\n```\n\n## Agent Recommendations\n\n### Primary Development Agents\n- **grok**: General coding and planning\n- **sonnet**: General coding and planning  \n- **supernova**: General coding and planning\n- **qwen3**: General coding and planning\n\n### Specialized Agents\n- **architect**: Architectural guidance and design decisions\n- **security-auditor**: Security audits and vulnerability identification\n- **review**: Code quality and best practices review\n\n## Key Benefits\n\n1. **Structured Process**: Clear separation between planning and execution\n2. **Quality Assurance**: Percentage-based validation ensures high-quality output\n3. **Architectural Integrity**: Research-backed architectural decisions\n4. **Iterative Improvement**: Automatic refinement until 90%+ quality achieved\n5. **Traceability**: Clear documentation trail from task to implementation\n6. **Separation of Concerns**: Architecture, planning, and implementation are handled separately\n7. **Reliability**: Retry limits, error handling, and no infinite loops 🛡️\n8. **Flexibility**: Dynamic agent selection with customizable combinations\n\n## Best Practices\n\n1. **Task Creation**: Be specific about requirements and include architectural considerations\n2. **Agent Selection**: Use different agents for planning (@agent_1) and review (@agent_2) to get diverse perspectives\n3. **Quality Standards**: The 90% score threshold ensures high-quality deliverables\n4. **Documentation**: Each stage produces comprehensive documentation for future reference\n5. **No Premature Commits**: Review all changes before committing to maintain code quality\n6. **Feature vs Task**: Use feature workflow for complex projects with 3+ related tasks\n7. **Error Handling**: Monitor retry counts and check for graceful failure handling\n8. **Quality Tracking**: Review percentage breakdowns to understand specific improvement areas\n\n## File Structure\n\n```\nproject/\n├── .feature/                      # Feature workflow\n│   ├── {id}-{title}.md           # Feature specifications\n│   ├── arch_{id}.md              # Essential architecture\n│   ├── arch_{id}_research.md     # Detailed research (optional)\n│   └── {id}-decomposition.md     # Task breakdown\n├── .task/\n│   └── {id}-{title}.md           # Individual tasks\n├── .plan/\n│   ├── arch_{id}.md              # Task-specific architecture\n│   └── {id}.md                   # Implementation plans\n├── .cache/                        # Support files\n│   └── arch_migration.md         # Migration guide\n├── templates/                     # Architecture templates\n│   ├── arch_essential.md         # Essential arch template\n│   └── arch_research.md          # Research template\n└── [your code files]             # Implementation results\n```\n\n---\n\nThis framework ensures systematic, high-quality feature development through coordinated agent collaboration while maintaining architectural integrity and code quality standards.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpigmej%2Fopencode_config","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpigmej%2Fopencode_config","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpigmej%2Fopencode_config/lists"}