[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MXS2ABZB9G6H95BVCGZ3K3":3,"skill-same:01M6MXS2ABZB9G6H95BVCGZ3K3":199},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":6,"ownerHtmlUrl":54,"ownerBlogUrl":6,"ownerDescription":55,"updatedAt":56,"install":57,"resources":70},"01M6MXS2ABZB9G6H95BVCGZ3K3","claw-conductor",null,"Always-on autonomous development orchestrator with intelligent triage. Auto-detects Discord channels, routes to project workspaces, triages simple vs development requests, decomposes complex tasks, routes to optimal AI models, executes in parallel, and consolidates results.","Kernel8901\u002Fai-agent-skills-classification","Kernel8901","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F72647553?v=4","01M6KV388J2H5PNC8D8P940BJD","01M6KV38CS5B3AY63RKZB410YM",7,"MIT","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"claw conductor","claw","conductor","workspaces","requests","with","channels","ai models","project","simple","models","discord","auto","orchestrator","autonomous","results","parallel","development","tasks","triage","2026-11-10 10:12:30",false,"---\nname: claw-conductor\ndescription: Always-on autonomous development orchestrator with intelligent triage. Auto-detects Discord channels, routes to project workspaces, triages simple vs development requests, decomposes complex tasks, routes to optimal AI models, executes in parallel, and consolidates results.\nversion: 2.1.0\n---\n\n# Claw Conductor v2.1\n\n**Your always-on development assistant - handles everything from quick questions to full project builds.**\n\nClaw Conductor is an intelligent orchestration layer that:\n\n- ? **Always-On**: Handles every message automatically (no need to invoke)\n- ? **Smart Triage**: Detects simple questions vs development tasks\n- ? **Discord-Aware**: Auto-maps channels to project workspaces\n- ? **Multi-Model**: Routes tasks to optimal AI based on capabilities\n- ⚡ **Parallel Execution**: Builds complete projects efficiently\n\n## ? How It Works\n\n**Automatic Flow:**\n1. Message arrives in Discord channel (e.g., #scientific-calculator)\n2. Claw-conductor detects channel → maps to `\u002Froot\u002Fprojects\u002Fscientific-calculator`\n3. Triages request: Simple question or development task?\n4. **If Simple**: Quick response from fast model with project context\n5. **If Development**: Full orchestration - decompose, route, execute, consolidate\n\n**You never need to explicitly invoke it** - it handles everything automatically!\n\n## ? Usage Examples\n\n**Simple Questions** (fast response):\n```\nUser: What files are in this project?\nConductor: ? Simple response mode\n          [Lists files from \u002Froot\u002Fprojects\u002Fscientific-calculator]\n\nUser: How does the calculator work?\nConductor: ? Simple response mode\n          [Explains architecture with project context]\n```\n\n**Development Tasks** (full orchestration):\n```\nUser: Build a scientific calculator with trig functions\nConductor: ? Development mode - full orchestration\n          [Decomposes into tasks, routes to models, executes in parallel]\n\nUser: Fix the bug in the calculation logic\nConductor: ? Development mode - full orchestration\n          [Analyzes code, creates fix, tests, commits]\n```\n\n**User Overrides**:\n```\nUser: !simple Build a calculator\nConductor: ? Simple response mode\n          [Gives advice\u002Fexplanation instead of building]\n\nUser: !dev What files exist?\nConductor: ? Development mode - full orchestration\n          [Treats as development task - maybe creates file listing tool]\n```\n\n## ? Skill Invocation (For OpenClaw Agent)\n\n**NEW: Always-On Mode (Recommended)**\n\nConfigure this skill as the default handler for Discord channels in \"Active Projects\" category:\n\n```python\n# In OpenClaw agent configuration\nfrom orchestrator import Orchestrator\n\norchestrator = Orchestrator()\n\n# Handle ALL messages through conductor\nresult = orchestrator.handle_message(\n    request=user_message,\n    channel_id=discord_channel_id,\n    channel_name=discord_channel_name\n)\n```\n\n**Legacy: Explicit Invocation**\n\nWhen this skill is invoked with a development request, execute the following:\n\n1. **Extract the request** from the user's message (everything after \"use claw-conductor to\")\n\n2. **Determine project name** from the request or generate one from keywords\n\n3. **Execute the orchestrator** using Python:\n   ```bash\n   cd ~\u002F.openclaw\u002Fskills\u002Fclaw-conductor\u002Fscripts\n   python3 -c \"\n   from orchestrator import Orchestrator\n   import sys\n\n   orchestrator = Orchestrator()\n\n   request = '''[USER'S REQUEST HERE]'''\n   project_name = '[PROJECT-NAME]'  # e.g., 'calculator-app', 'todo-app', 'blog-site'\n\n   # Get GitHub user from config\n   github_user = orchestrator.config.get('github_user')\n\n   result = orchestrator.execute_request(\n       request=request,\n       project_name=project_name,\n       github_user=github_user\n   )\n\n   # Report results back to Discord\n   if result['success']:\n       print(f\\\"✅ Project '{project_name}' completed successfully!\\\")\n       print(f\\\"? {result['tasks_completed']} tasks completed\\\")\n       if github_user:\n           print(f\\\"? GitHub: https:\u002F\u002Fgithub.com\u002F{github_user}\u002F{project_name}\\\")\n       print(f\\\"? Workspace: {result.get('workspace', '\u002Froot\u002Fprojects\u002F' + project_name)}\\\")\n   else:\n       print(f\\\"❌ Project failed: {result.get('error', 'Unknown error')}\\\")\n       sys.exit(1)\n   \"\n   ```\n\n4. **Report progress** to Discord during execution:\n   - Announce task decomposition results\n   - Report task routing decisions\n   - Update on parallel execution progress\n   - Share final results with GitHub link\n\n**Example Invocation:**\nUser says: `@OpenClaw use claw-conductor to build a calculator app`\n\nYou execute:\n- Request: \"build a calculator app\"\n- Project name: \"calculator-app\"\n- Run orchestrator with these parameters\n\n---\n\n## What's New in v2.1\n\n? **AI-Powered Decomposition**: Intelligently analyzes complex requests using your best AI model (auto-selected or configured)\n? **Full Orchestration**: Decomposes complex requests → Routes subtasks → Executes in parallel → Consolidates results\n⚡ **Parallel Execution**: Up to 5 tasks running concurrently across multiple projects\n? **Project Management**: Automatic workspace creation, git initialization, and GitHub integration\n? **Dependency-Aware**: Respects task dependencies and file conflicts\n? **Auto-Consolidation**: Merges results, runs tests, commits to git, pushes to GitHub\n\n---\n\n## Quick Start\n\n### Installation\n\nIn OpenClaw:\n```bash\ncd ~\u002F.openclaw\u002Fskills\ngit clone https:\u002F\u002Fgithub.com\u002Fjohnsonfarmsus\u002Fclaw-conductor.git\ncd claw-conductor\n.\u002Fscripts\u002Fsetup.sh\n```\n\n### First-Time Setup\n\n```bash\n.\u002Fscripts\u002Fsetup.sh\n```\n\nThis creates your personalized agent-registry.json with:\n- Your AI model configurations\n- Cost tracking (free vs paid)\n- Capability ratings per model\n- Routing preferences\n\n### Usage\n\n**Simple request:**\n```\n@OpenClaw use claw-conductor to build a calculator app\n```\n\n**Complex request:**\n```\n@OpenClaw use claw-conductor to build a towing dispatch system with:\n- Customer portal for requesting service\n- Driver dashboard for accepting jobs\n- Admin panel for managing users\n- Real-time location tracking\n- Payment integration\n```\n\n---\n\n## How It Works\n\n### Complete Workflow\n\n```\nDiscord Request\n    ↓\n1. Task Decomposition\n   • Analyzes request complexity\n   • Breaks into independent subtasks\n   • Assigns category & complexity to each\n   • Builds dependency graph\n    ↓\n2. Intelligent Routing\n   • Scores each model for each task (0-100)\n   • Routes to best match based on capabilities\n   • Considers cost optimization\n    ↓\n3. Project Initialization\n   • Creates \u002Froot\u002Fprojects\u002F{name}\u002F\n   • Initializes git repository\n   • Creates GitHub repo (if configured)\n   • Sets up workspace\n    ↓\n4. Parallel Execution\n   • Spawns up to 5 tasks simultaneously\n   • Respects dependencies (database before auth)\n   • Avoids file conflicts (same files sequential)\n   • Reports progress to Discord\n    ↓\n5. Result Consolidation\n   • Merges all task outputs\n   • Resolves file conflicts\n   • Runs tests (if present)\n   • Commits to git\n   • Pushes to GitHub\n    ↓\nDiscord Completion Report\n```\n\n### Example: Dispatch System\n\n**Request:**\n```\nBuild a towing dispatch system with customer portal,\ndriver dashboard, admin panel, and real-time tracking\n```\n\n**Decomposition:**\n```\nTask 1: Database schema (database-operations, complexity: 4)\nTask 2: Authentication system (security-fixes, complexity: 4)\nTask 3: Customer portal UI (frontend-development, complexity: 3)\nTask 4: Driver dashboard UI (frontend-development, complexity: 3)\nTask 5: Admin panel UI (frontend-development, complexity: 3)\nTask 6: REST API endpoints (api-development, complexity: 3)\nTask 7: Real-time tracking (performance-optimization, complexity: 5)\nTask 8: Unit tests (unit-test-generation, complexity: 2)\n```\n\n**Routing:**\n```\nTask 1 → Mistral Devstral (score: 92, best for database)\nTask 2 → Mistral Devstral (score: 88, security expert)\nTask 3 → Mistral Devstral (score: 95, frontend expert)\nTask 4 → Mistral Devstral (score: 95, frontend expert)\nTask 5 → Mistral Devstral (score: 95, frontend expert)\nTask 6 → Llama 3.3 70B (score: 87, API specialist)\nTask 7 → Mistral Devstral (score: 78, fallback - needs Claude ideally)\nTask 8 → Llama 3.3 70B (score: 95, test generation expert)\n```\n\n**Execution:**\n```\nParallel execution plan:\nWorker 1: Task 1 (Database) → Mistral\nWorker 2: Task 3 (Customer UI) → Devstral\nWorker 3: Task 4 (Driver UI) → Devstral\nWorker 4: Task 5 (Admin UI) → Devstral\nWorker 5: Task 6 (API) → Llama\n\nAfter Task 1 completes:\nWorker 1: Task 2 (Auth - depends on DB) → Mistral\n\nAfter all code complete:\nWorker 1: Task 8 (Tests) → Llama\n```\n\n**Result:**\n```\n✅ All 8 tasks completed in 47 minutes\n? Committed to git with 8 changes\n? Pushed to GitHub repository\n? Project ready for deployment\n```\n\n---\n\n## Scoring Algorithm\n\nEach model is scored 0-100 for each task:\n\n```python\nscore = (\n    (rating \u002F 5.0) * 50 +              # Model capability (0-50 pts)\n    (1 - complexity\u002F5.0) * 40 +        # Complexity fit (0-40 pts)\n    (experience \u002F 100) * 10 +          # Experience (0-10 pts)\n    cost_factor * 10                   # Cost (0-10 pts)\n)\n```\n\n**Hard Ceiling:** Models cannot handle tasks above their `max_complexity` rating.\n\n### Scoring Example\n\n**Task:** Backend API development (complexity: 4)\n\n| Model | Capability | Complexity Fit | Experience | Cost | Total |\n|-------|------------|----------------|------------|------|-------|\n| Mistral Devstral | 4★ (40pts) | Can handle 4 (40pts) | 0 (0pts) | Free (10pts) | **90\u002F100** |\n| Llama 3.3 70B | 4★ (40pts) | Can handle 4 (40pts) | 2 tasks (2pts) | Free (10pts) | **92\u002F100** ✅ |\n| Perplexity | N\u002FA | Cannot handle backend | - | - | **0\u002F100** |\n\nWinner: **Llama 3.3 70B** (higher experience)\n\n---\n\n## Configuration\n\n### Agent Registry Structure\n\n`config\u002Fagent-registry.json`:\n```json\n{\n  \"version\": \"1.0.0\",\n  \"user_config\": {\n    \"cost_tracking_enabled\": true,\n    \"prefer_free_when_equal\": true,\n    \"max_parallel_tasks\": 5,\n    \"default_complexity_if_unknown\": 3,\n    \"fallback\": {\n      \"enabled\": true,\n      \"retry_delay_seconds\": 2,\n      \"track_failures\": true,\n      \"penalize_failures\": true,\n      \"failure_penalty_points\": 5\n    }\n  },\n  \"agents\": {\n    \"mistral-devstral-2512\": {\n      \"model_id\": \"mistral\u002Fdevstral-2512\",\n      \"provider\": \"mistral\",\n      \"context_window\": 256000,\n      \"enabled\": true,\n      \"user_cost\": {\n        \"type\": \"free-tier\",\n        \"input_cost_per_million\": 0,\n        \"output_cost_per_million\": 0\n      },\n      \"capabilities\": {\n        \"frontend-development\": {\n          \"rating\": 5,\n          \"max_complexity\": 5,\n          \"notes\": \"Expert - near-parity with Claude\"\n        },\n        \"multi-file-refactoring\": {\n          \"rating\": 5,\n          \"max_complexity\": 5,\n          \"notes\": \"Expert - designed for 50+ file changes\"\n        }\n      }\n    }\n  }\n}\n```\n\n### Fallback Strategy\n\nConservative fallback (user-configurable):\n1. Try primary model (attempt 1)\n2. Try primary model (attempt 2)\n3. If both fail → Try first runner-up (attempt 3)\n4. Try first runner-up (attempt 4)\n5. If all fail → Give up, report to Discord\n\n**Why conservative?**\nPrevents cascading through irrelevant models that may not have capability for the task.\n\n---\n\n## Task Categories (23 Standard)\n\n- code-generation-new-features\n- bug-detection-fixes\n- multi-file-refactoring\n- unit-test-generation\n- debugging-complex-issues\n- api-development\n- security-vulnerability-detection\n- security-fixes\n- documentation-generation\n- code-review\n- frontend-development\n- backend-development\n- database-operations\n- codebase-exploration\n- dependency-management\n- legacy-modernization\n- error-correction\n- performance-optimization\n- test-coverage-analysis\n- algorithm-implementation\n- boilerplate-generation\n\n---\n\n## Advanced Features\n\n### Multi-Project Support\n\nHandle concurrent requests across different projects:\n\n```\nProject A: Dispatch System (3 tasks running)\nProject B: Calculator App (2 tasks running)\n────────────────────────────────────────────\nTotal: 5 concurrent tasks (at global limit)\n```\n\n### File Conflict Detection\n\nTasks touching the same files run sequentially:\n\n```\nTask 1: Modify src\u002Fapi\u002Fusers.js → Running\nTask 2: Modify src\u002Fapi\u002Fusers.js → Queued (waits for Task 1)\nTask 3: Modify src\u002Fui\u002Fdashboard.js → Running (independent)\n```\n\n### Dependency-Aware Scheduling\n\n```\nTask 1: Database schema → No deps, starts immediately\nTask 2: Auth system → Depends on Task 1, waits\nTask 3: Frontend UI → Depends on Task 2, waits\nTask 4: Tests → Depends on all, runs last\n```\n\n### Auto-Consolidation\n\nAfter all tasks complete:\n1. Check git status for conflicts\n2. Run tests (pytest, npm test, etc.)\n3. Commit with conventional commit message\n4. Push to GitHub (if configured)\n5. Report to Discord\n\n---\n\n## Examples\n\n### Simple Calculator\n\n```\n@OpenClaw use claw-conductor to build a calculator with:\n- Basic operations (add, subtract, multiply, divide)\n- Clean UI\n- Unit tests\n```\n\n**Result:**\n- 3 tasks (UI, logic, tests)\n- Completed in ~8 minutes\n- Pushed to GitHub\n\n### Towing Dispatch System\n\n```\n@OpenClaw use claw-conductor to build a dispatch system with:\n- Customer portal\n- Driver dashboard\n- Admin panel\n- Real-time tracking\n- Payment integration\n```\n\n**Result:**\n- 8 tasks across 3 models\n- Completed in ~45 minutes\n- Full working application\n\n### API with Documentation\n\n```\n@OpenClaw use claw-conductor to create a REST API for a blog with:\n- CRUD operations for posts\n- Authentication\n- Swagger documentation\n- Integration tests\n```\n\n**Result:**\n- 5 tasks (schema, auth, endpoints, docs, tests)\n- Completed in ~20 minutes\n- API-first design\n\n---\n\n## Troubleshooting\n\n### Task Decomposition Issues\n\n**Problem:** Request not decomposed correctly\n**Solution:** Be specific in request. Include keywords: \"database\", \"API\", \"frontend\", \"tests\"\n\n### Model Selection Issues\n\n**Problem:** Wrong model chosen for task\n**Solution:** Adjust capability ratings in `agent-registry.json`\n\n### Execution Failures\n\n**Problem:** Task fails with error\n**Solution:** Fallback tries primary 2x, runner-up 2x. Check error logs in `.claw-conductor\u002Fexecution-log.json`\n\n### Git Conflicts\n\n**Problem:** Consolidation fails due to conflicts\n**Solution:** Currently requires manual resolution. Future: AI-powered conflict resolution\n\n---\n\n## Roadmap\n\n- [x] Task decomposition (v2.0)\n- [x] Parallel execution (v2.0)\n- [x] Multi-project support (v2.0)\n- [x] Auto-consolidation (v2.0)\n- [ ] AI-powered decomposition (v2.1)\n- [ ] Discord progress updates (v2.1)\n- [ ] Conflict resolution with AI (v2.2)\n- [ ] Real-time task streaming (v2.2)\n- [ ] Web dashboard (v3.0)\n\n---\n\n## License\n\nGNU AGPL v3 - See LICENSE file\n\nCopyleft license requiring server-side source availability.\n\n---\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.\n\nPublished on ClawHub.ai: https:\u002F\u002Fclawhub.ai\u002Fskills\u002Fclaw-conductor\n\n---\n\n*Built with ❤️ by the Claw Conductor team*\n","# Claw Conductor v2.1\n\n**Your always-on development assistant - handles everything from quick questions to full project builds.**\n\nClaw Conductor is an intelligent orchestration layer that:\n\n- ? **Always-On**: Handles every message automatically (no need to invoke)\n- ? **Smart Triage**: Detects simple questions vs development tasks\n- ? **Discord-Aware**: Auto-maps channels to project workspaces\n- ? **Multi-Model**: Routes tasks to optimal AI based on capabilities\n- ⚡ **Parallel Execution**: Builds complete projects efficiently\n\n## ? How It Works\n\n**Automatic Flow:**\n1. Message arrives in Discord channel (e.g., #scientific-calculator)\n2. Claw-conductor detects channel → maps to `\u002Froot\u002Fprojects\u002Fscientific-calculator`\n3. Triages request: Simple question or development task?\n4. **If Simple**: Quick response from fast model with project context\n5. **If Development**: Full orchestration - decompose, route, execute, consolidate\n\n**You never need to explicitly invoke it** - it handles everything automatically!\n\n## ? Usage Examples\n\n**Simple Questions** (fast response):\n```\nUser: What files are in this project?\nConductor: ? Simple response mode\n          [Lists files from \u002Froot\u002Fprojects\u002Fscientific-calculator]\n\nUser: How does the calculator work?\nConductor: ? Simple response mode\n          [Explains architecture with project context]\n```\n\n**Development Tasks** (full orchestration):\n```\nUser: Build a scientific calculator with trig functions\nConductor: ? Development mode - full orchestration\n          [Decomposes into tasks, routes to models, executes in parallel]\n\nUser: Fix the bug in the calculation logic\nConductor: ? Development mode - full orchestration\n          [Analyzes code, creates fix, tests, commits]\n```\n\n**User Overrides**:\n```\nUser: !simple Build a calculator\nConductor: ? Simple response mode\n          [Gives advice\u002Fexplanation instead of building]\n\nUser: !dev What files exist?\nConductor: ? Development mode - full orchestration\n          [Treats as development task - maybe creates file listing tool]\n```\n\n## ? Skill Invocation (For OpenClaw Agent)\n\n**NEW: Always-On Mode (Recommended)**\n\nConfigure this skill as the default handler for Discord channels in \"Active Projects\" category:\n\n```python\n# In OpenClaw agent configuration\nfrom orchestrator import Orchestrator\n\norchestrator = Orchestrator()\n\n# Handle ALL messages through conductor\nresult = orchestrator.handle_message(\n    request=user_message,\n    channel_id=discord_channel_id,\n    channel_name=discord_channel_name\n)\n```\n\n**Legacy: Explicit Invocation**\n\nWhen this skill is invoked with a development request, execute the following:\n\n1. **Extract the request** from the user's message (everything after \"use claw-conductor to\")\n\n2. **Determine project name** from the request or generate one from keywords\n\n3. **Execute the orchestrator** using Python:\n   ```bash\n   cd ~\u002F.openclaw\u002Fskills\u002Fclaw-conductor\u002Fscripts\n   python3 -c \"\n   from orchestrator import Orchestrator\n   import sys\n\n   orchestrator = Orchestrator()\n\n   request = '''[USER'S REQUEST HERE]'''\n   project_name = '[PROJECT-NAME]'  # e.g., 'calculator-app', 'todo-app', 'blog-site'\n\n   # Get GitHub user from config\n   github_user = orchestrator.config.get('github_user')\n\n   result = orchestrator.execute_request(\n       request=request,\n       project_name=project_name,\n       github_user=github_user\n   )\n\n   # Report results back to Discord\n   if result['success']:\n       print(f\\\"✅ Project '{project_name}' completed successfully!\\\")\n       print(f\\\"? {result['tasks_completed']} tasks completed\\\")\n       if github_user:\n           print(f\\\"? GitHub: https:\u002F\u002Fgithub.com\u002F{github_user}\u002F{project_name}\\\")\n       print(f\\\"? Workspace: {result.get('workspace', '\u002Froot\u002Fprojects\u002F' + project_name)}\\\")\n   else:\n       print(f\\\"❌ Project failed: {result.get('error', 'Unknown error')}\\\")\n       sys.exit(1)\n   \"\n   ```\n\n4. **Report progress** to Discord during execution:\n   - Announce task decomposition results\n   - Report task routing decisions\n   - Update on parallel execution progress\n   - Share final results with GitHub link\n\n**Example Invocation:**\nUser says: `@OpenClaw use claw-conductor to build a calculator app`\n\nYou execute:\n- Request: \"build a calculator app\"\n- Project name: \"calculator-app\"\n- Run orchestrator with these parameters\n\n---\n\n## What's New in v2.1\n\n? **AI-Powered Decomposition**: Intelligently analyzes complex requests using your best AI model (auto-selected or configured)\n? **Full Orchestration**: Decomposes complex requests → Routes subtasks → Executes in parallel → Consolidates results\n⚡ **Parallel Execution**: Up to 5 tasks running concurrently across multiple projects\n? **Project Management**: Automatic workspace creation, git initialization, and GitHub integration\n? **Dependency-Aware**: Respects task dependencies and file conflicts\n? **Auto-Consolidation**: Merges results, runs tests, commits to git, pushes to GitHub\n\n---\n\n## Quick Start\n\n### Installation\n\nIn OpenClaw:\n```bash\ncd ~\u002F.openclaw\u002Fskills\ngit clone https:\u002F\u002Fgithub.com\u002Fjohnsonfarmsus\u002Fclaw-conductor.git\ncd claw-conductor\n.\u002Fscripts\u002Fsetup.sh\n```\n\n### First-Time Setup\n\n```bash\n.\u002Fscripts\u002Fsetup.sh\n```\n\nThis creates your personalized agent-registry.json with:\n- Your AI model configurations\n- Cost tracking (free vs paid)\n- Capability ratings per model\n- Routing preferences\n\n### Usage\n\n**Simple request:**\n```\n@OpenClaw use claw-conductor to build a calculator app\n```\n\n**Complex request:**\n```\n@OpenClaw use claw-conductor to build a towing dispatch system with:\n- Customer portal for requesting service\n- Driver dashboard for accepting jobs\n- Admin panel for managing users\n- Real-time location tracking\n- Payment integration\n```\n\n---\n\n## How It Works\n\n### Complete Workflow\n\n```\nDiscord Request\n    ↓\n1. Task Decomposition\n   • Analyzes request complexity\n   • Breaks into independent subtasks\n   • Assigns category & complexity to each\n   • Builds dependency graph\n    ↓\n2. Intelligent Routing\n   • Scores each model for each task (0-100)\n   • Routes to best match based on capabilities\n   • Considers cost optimization\n    ↓\n3. Project Initialization\n   • Creates \u002Froot\u002Fprojects\u002F{name}\u002F\n   • Initializes git repository\n   • Creates GitHub repo (if configured)\n   • Sets up workspace\n    ↓\n4. Parallel Execution\n   • Spawns up to 5 tasks simultaneously\n   • Respects dependencies (database before auth)\n   • Avoids file conflicts (same files sequential)\n   • Reports progress to Discord\n    ↓\n5. Result Consolidation\n   • Merges all task outputs\n   • Resolves file conflicts\n   • Runs tests (if present)\n   • Commits to git\n   • Pushes to GitHub\n    ↓\nDiscord Completion Report\n```\n\n### Example: Dispatch System\n\n**Request:**\n```\nBuild a towing dispatch system with customer portal,\ndriver dashboard, admin panel, and real-time tracking\n```\n\n**Decomposition:**\n```\nTask 1: Database schema (database-operations, complexity: 4)\nTask 2: Authentication system (security-fixes, complexity: 4)\nTask 3: Customer portal UI (frontend-development, complexity: 3)\nTask 4: Driver dashboard UI (frontend-development, complexity: 3)\nTask 5: Admin panel UI (frontend-development, complexity: 3)\nTask 6: REST API endpoints (api-development, complexity: 3)\nTask 7: Real-time tracking (performance-optimization, complexity: 5)\nTask 8: Unit tests (unit-test-generation, complexity: 2)\n```\n\n**Routing:**\n```\nTask 1 → Mistral Devstral (score: 92, best for database)\nTask 2 → Mistral Devstral (score: 88, security expert)\nTask 3 → Mistral Devstral (score: 95, frontend expert)\nTask 4 → Mistral Devstral (score: 95, frontend expert)\nTask 5 → Mistral Devstral (score: 95, frontend expert)\nTask 6 → Llama 3.3 70B (score: 87, API specialist)\nTask 7 → Mistral Devstral (score: 78, fallback - needs Claude ideally)\nTask 8 → Llama 3.3 70B (score: 95, test generation expert)\n```\n\n**Execution:**\n```\nParallel execution plan:\nWorker 1: Task 1 (Database) → Mistral\nWorker 2: Task 3 (Customer UI) → Devstral\nWorker 3: Task 4 (Driver UI) → Devstral\nWorker 4: Task 5 (Admin UI) → Devstral\nWorker 5: Task 6 (API) → Llama\n\nAfter Task 1 completes:\nWorker 1: Task 2 (Auth - depends on DB) → Mistral\n\nAfter all code complete:\nWorker 1: Task 8 (Tests) → Llama\n```\n\n**Result:**\n```\n✅ All 8 tasks completed in 47 minutes\n? Committed to git with 8 changes\n? Pushed to GitHub repository\n? Project ready for deployment\n```\n\n---\n\n## Scoring Algorithm\n\nEach model is scored 0-100 for each task:\n\n```python\nscore = (\n    (rating \u002F 5.0) * 50 +              # Model capability (0-50 pts)\n    (1 - complexity\u002F5.0) * 40 +        # Complexity fit (0-40 pts)\n    (experience \u002F 100) * 10 +          # Experience (0-10 pts)\n    cost_factor * 10                   # Cost (0-10 pts)\n)\n```\n\n**Hard Ceiling:** Models cannot handle tasks above their `max_complexity` rating.\n\n### Scoring Example\n\n**Task:** Backend API development (complexity: 4)\n\n| Model | Capability | Complexity Fit | Experience | Cost | Total |\n|-------|------------|----------------|------------|------|-------|\n| Mistral Devstral | 4★ (40pts) | Can handle 4 (40pts) | 0 (0pts) | Free (10pts) | **90\u002F100** |\n| Llama 3.3 70B | 4★ (40pts) | Can handle 4 (40pts) | 2 tasks (2pts) | Free (10pts) | **92\u002F100** ✅ |\n| Perplexity | N\u002FA | Cannot handle backend | - | - | **0\u002F100** |\n\nWinner: **Llama 3.3 70B** (higher experience)\n\n---\n\n## Configuration\n\n### Agent Registry Structure\n\n`config\u002Fagent-registry.json`:\n```json\n{\n  \"version\": \"1.0.0\",\n  \"user_config\": {\n    \"cost_tracking_enabled\": true,\n    \"prefer_free_when_equal\": true,\n    \"max_parallel_tasks\": 5,\n    \"default_complexity_if_unknown\": 3,\n    \"fallback\": {\n      \"enabled\": true,\n      \"retry_delay_seconds\": 2,\n      \"track_failures\": true,\n      \"penalize_failures\": true,\n      \"failure_penalty_points\": 5\n    }\n  },\n  \"agents\": {\n    \"mistral-devstral-2512\": {\n      \"model_id\": \"mistral\u002Fdevstral-2512\",\n      \"provider\": \"mistral\",\n      \"context_window\": 256000,\n      \"enabled\": true,\n      \"user_cost\": {\n        \"type\": \"free-tier\",\n        \"input_cost_per_million\": 0,\n        \"output_cost_per_million\": 0\n      },\n      \"capabilities\": {\n        \"frontend-development\": {\n          \"rating\": 5,\n          \"max_complexity\": 5,\n          \"notes\": \"Expert - near-parity with Claude\"\n        },\n        \"multi-file-refactoring\": {\n          \"rating\": 5,\n          \"max_complexity\": 5,\n          \"notes\": \"Expert - designed for 50+ file changes\"\n        }\n      }\n    }\n  }\n}\n```\n\n### Fallback Strategy\n\nConservative fallback (user-configurable):\n1. Try primary model (attempt 1)\n2. Try primary model (attempt 2)\n3. If both fail → Try first runner-up (attempt 3)\n4. Try first runner-up (attempt 4)\n5. If all fail → Give up, report to Discord\n\n**Why conservative?**\nPrevents cascading through irrelevant models that may not have capability for the task.\n\n---\n\n## Task Categories (23 Standard)\n\n- code-generation-new-features\n- bug-detection-fixes\n- multi-file-refactoring\n- unit-test-generation\n- debugging-complex-issues\n- api-development\n- security-vulnerability-detection\n- security-fixes\n- documentation-generation\n- code-review\n- frontend-development\n- backend-development\n- database-operations\n- codebase-exploration\n- dependency-management\n- legacy-modernization\n- error-correction\n- performance-optimization\n- test-coverage-analysis\n- algorithm-implementation\n- boilerplate-generation\n\n---\n\n## Advanced Features\n\n### Multi-Project Support\n\nHandle concurrent requests across different projects:\n\n```\nProject A: Dispatch System (3 tasks running)\nProject B: Calculator App (2 tasks running)\n────────────────────────────────────────────\nTotal: 5 concurrent tasks (at global limit)\n```\n\n### File Conflict Detection\n\nTasks touching the same files run sequentially:\n\n```\nTask 1: Modify src\u002Fapi\u002Fusers.js → Running\nTask 2: Modify src\u002Fapi\u002Fusers.js → Queued (waits for Task 1)\nTask 3: Modify src\u002Fui\u002Fdashboard.js → Running (independent)\n```\n\n### Dependency-Aware Scheduling\n\n```\nTask 1: Database schema → No deps, starts immediately\nTask 2: Auth system → Depends on Task 1, waits\nTask 3: Frontend UI → Depends on Task 2, waits\nTask 4: Tests → Depends on all, runs last\n```\n\n### Auto-Consolidation\n\nAfter all tasks complete:\n1. Check git status for conflicts\n2. Run tests (pytest, npm test, etc.)\n3. Commit with conventional commit message\n4. Push to GitHub (if configured)\n5. Report to Discord\n\n---\n\n## Examples\n\n### Simple Calculator\n\n```\n@OpenClaw use claw-conductor to build a calculator with:\n- Basic operations (add, subtract, multiply, divide)\n- Clean UI\n- Unit tests\n```\n\n**Result:**\n- 3 tasks (UI, logic, tests)\n- Completed in ~8 minutes\n- Pushed to GitHub\n\n### Towing Dispatch System\n\n```\n@OpenClaw use claw-conductor to build a dispatch system with:\n- Customer portal\n- Driver dashboard\n- Admin panel\n- Real-time tracking\n- Payment integration\n```\n\n**Result:**\n- 8 tasks across 3 models\n- Completed in ~45 minutes\n- Full working application\n\n### API with Documentation\n\n```\n@OpenClaw use claw-conductor to create a REST API for a blog with:\n- CRUD operations for posts\n- Authentication\n- Swagger documentation\n- Integration tests\n```\n\n**Result:**\n- 5 tasks (schema, auth, endpoints, docs, tests)\n- Completed in ~20 minutes\n- API-first design\n\n---\n\n## Troubleshooting\n\n### Task Decomposition Issues\n\n**Problem:** Request not decomposed correctly\n**Solution:** Be specific in request. Include keywords: \"database\", \"API\", \"frontend\", \"tests\"\n\n### Model Selection Issues\n\n**Problem:** Wrong model chosen for task\n**Solution:** Adjust capability ratings in `agent-registry.json`\n\n### Execution Failures\n\n**Problem:** Task fails with error\n**Solution:** Fallback tries primary 2x, runner-up 2x. Check error logs in `.claw-conductor\u002Fexecution-log.json`\n\n### Git Conflicts\n\n**Problem:** Consolidation fails due to conflicts\n**Solution:** Currently requires manual resolution. Future: AI-powered conflict resolution\n\n---\n\n## Roadmap\n\n- [x] Task decomposition (v2.0)\n- [x] Parallel execution (v2.0)\n- [x] Multi-project support (v2.0)\n- [x] Auto-consolidation (v2.0)\n- [ ] AI-powered decomposition (v2.1)\n- [ ] Discord progress updates (v2.1)\n- [ ] Conflict resolution with AI (v2.2)\n- [ ] Real-time task streaming (v2.2)\n- [ ] Web dashboard (v3.0)\n\n---\n\n## License\n\nGNU AGPL v3 - See LICENSE file\n\nCopyleft license requiring server-side source availability.\n\n---\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.\n\nPublished on ClawHub.ai: https:\u002F\u002Fclawhub.ai\u002Fskills\u002Fclaw-conductor\n\n---\n\n*Built with ❤️ by the Claw Conductor team*\n",{"name":5,"version":48,"description":7},"2.1.0","main","f4cb2d49f7667e79ef7804f50396d5aaffdf1f56","人工智能与大型语言模型\u002Fclaw-conductor",1,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification","https:\u002F\u002Fgithub.com\u002FKernel8901","a little little little dobe in life but science geek","2026-04-05 16:43:56",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add Kernel8901\u002Fai-agent-skills-classification --skill claw-conductor",{"tool":63,"command":64},"bunx","bunx skills add Kernel8901\u002Fai-agent-skills-classification --skill claw-conductor",{"tool":66,"command":67},"pnpm","pnpm dlx skills add Kernel8901\u002Fai-agent-skills-classification --skill claw-conductor","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FSKILL.md",true,[71,77,81,85,89,93,97,101,105,109,113,117,121,125,129,135,139,143,147,151,155,159,163,167,171,175,179,183,187,191,195],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"_meta.json","REFERENCE","参考文档",1170,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002F_meta.json",{"path":78,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":79,"sourceHtmlUrl":80},"config\u002Fagent-registry.example.json",3251,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fagent-registry.example.json",{"path":82,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":83,"sourceHtmlUrl":84},"config\u002Fchannel-workspace-map.json",851,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fchannel-workspace-map.json",{"path":86,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":87,"sourceHtmlUrl":88},"config\u002Fconductor-config.json",1311,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fconductor-config.json",{"path":90,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":91,"sourceHtmlUrl":92},"config\u002Fdefaults\u002Fclaude-sonnet-4.5.json",3919,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002Fclaude-sonnet-4.5.json",{"path":94,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":95,"sourceHtmlUrl":96},"config\u002Fdefaults\u002Fdeepseek-v3.json",1469,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002Fdeepseek-v3.json",{"path":98,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":99,"sourceHtmlUrl":100},"config\u002Fdefaults\u002Fgemini-2.0-flash.json",1639,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002Fgemini-2.0-flash.json",{"path":102,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":103,"sourceHtmlUrl":104},"config\u002Fdefaults\u002Fgpt-4-turbo.json",2159,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002Fgpt-4-turbo.json",{"path":106,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":107,"sourceHtmlUrl":108},"config\u002Fdefaults\u002Fqwen-2.5-72b.json",1563,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002Fqwen-2.5-72b.json",{"path":110,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":111,"sourceHtmlUrl":112},"config\u002Fdefaults\u002FREADME.md",3076,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Fdefaults\u002FREADME.md",{"path":114,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":115,"sourceHtmlUrl":116},"config\u002Ftask-categories.json",12554,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fconfig\u002Ftask-categories.json",{"path":118,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":119,"sourceHtmlUrl":120},"CONTRIBUTING.md",7754,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FCONTRIBUTING.md",{"path":122,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":123,"sourceHtmlUrl":124},"DEPLOYMENT.md",5991,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FDEPLOYMENT.md",{"path":126,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":127,"sourceHtmlUrl":128},"docs\u002Forchestration-design.md",16789,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fdocs\u002Forchestration-design.md",{"path":130,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":133,"sourceHtmlUrl":134},"examples\u002Fcomplex-feature.py","SCRIPT","脚本",4964,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fexamples\u002Fcomplex-feature.py",{"path":136,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":137,"sourceHtmlUrl":138},"examples\u002Ffallback-routing.py",9273,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fexamples\u002Ffallback-routing.py",{"path":140,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":141,"sourceHtmlUrl":142},"examples\u002FREADME.md",5782,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fexamples\u002FREADME.md",{"path":144,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":145,"sourceHtmlUrl":146},"examples\u002Fsimple-bug-fix.py",3241,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fexamples\u002Fsimple-bug-fix.py",{"path":148,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":149,"sourceHtmlUrl":150},"IMPLEMENTATION-STATUS.md",8151,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FIMPLEMENTATION-STATUS.md",{"path":152,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":153,"sourceHtmlUrl":154},"README.md",15389,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FREADME.md",{"path":156,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":157,"sourceHtmlUrl":158},"scripts\u002Fconsolidator.py",10841,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fconsolidator.py",{"path":160,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":161,"sourceHtmlUrl":162},"scripts\u002Fdecomposer.py",16127,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fdecomposer.py",{"path":164,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":165,"sourceHtmlUrl":166},"scripts\u002Fdiscord_integration.py",9607,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fdiscord_integration.py",{"path":168,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":169,"sourceHtmlUrl":170},"scripts\u002Forchestrator.py",18135,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Forchestrator.py",{"path":172,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":173,"sourceHtmlUrl":174},"scripts\u002Fproject_manager.py",7827,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fproject_manager.py",{"path":176,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":177,"sourceHtmlUrl":178},"scripts\u002Frouter.py",17698,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Frouter.py",{"path":180,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":181,"sourceHtmlUrl":182},"scripts\u002Fsetup.sh",13212,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fsetup.sh",{"path":184,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":185,"sourceHtmlUrl":186},"scripts\u002Ftest_decomposer.py",8824,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Ftest_decomposer.py",{"path":188,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":189,"sourceHtmlUrl":190},"scripts\u002Fupdate-capability.py",9900,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fupdate-capability.py",{"path":192,"resourceType":131,"resourceTypeDesc":132,"sizeBytes":193,"sourceHtmlUrl":194},"scripts\u002Fworker_pool.py",14804,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002Fscripts\u002Fworker_pool.py",{"path":196,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":197,"sourceHtmlUrl":198},"SETUP.md",7294,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fclaw-conductor\u002FSETUP.md",[200,224,256,267,277,307],{"skillNo":201,"name":202,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":204,"categories":205,"keywords":206,"publishedAt":43},"01M6Q5R6WZ06RRB9RQTW54B34B","zyfai","Earn yield on any Ethereum wallet on Base, Arbitrum, and Plasma. Use when a user wants passive DeFi yield on their funds. Deploys a non-custodial subaccount (Safe) linked to their EOA, enables automated yield optimization, and lets them deposit\u002Fwithdraw anytime.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzyfai-sdk\u002FSKILL.md",[],[202,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223],"safe","ethereum","defi","Optimization","user","use","Wallet","base","deposit","withdraw","yield","arbitrum","non-custodial","passive","defi yield","earn","yield optimization",{"skillNo":225,"name":226,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":227,"categories":228,"keywords":238,"publishedAt":43},"01M6Q5R337B5JSNER4P4YS03TN","zerion-api","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api\u002FSKILL.md",[229,232,233],{"categoryNo":230,"categoryName":231,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":234,"categoryName":235,"taxonomyType":236,"taxonomyTypeDesc":237},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[239,240,241,242,243,244,245,209,246,213,247,248,249,250,251,252,253,254,255],"api","zerion api","zerion","MCP","mcp server","server","blockchain","transactions","crypto wallet","crypto","evm","solana","query","across","positions","token","prices",{"skillNo":257,"name":258,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":259,"categories":260,"keywords":264,"publishedAt":43},"01M6Q5QZFYTDAZMZ27BN5VZXAR","zerion-api-skill","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api-skill\u002FSKILL.md",[261,262,263],{"categoryNo":230,"categoryName":231,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":234,"categoryName":235,"taxonomyType":236,"taxonomyTypeDesc":237},[239,265,266,240,241,242,243,244,245,209,246,213,248,249,250,251,252,253,254,255],"api skill","zerion api skill",{"skillNo":268,"name":269,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":270,"categories":271,"keywords":275,"publishedAt":43},"01M6Q5QW17QW98RA3FE45E4RD8","zerion-api-mcp","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api-mcp\u002FSKILL.md",[272,273,274],{"categoryNo":230,"categoryName":231,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":234,"categoryName":235,"taxonomyType":236,"taxonomyTypeDesc":237},[242,239,276,240,241,243,244,209,246,213,247,248,249,250,251,252,253,254,255],"zerion api mcp",{"skillNo":278,"name":279,"nameZh":6,"description":280,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":281,"categories":282,"keywords":286,"publishedAt":43},"01M6Q5QQYZB7A501BDEQWKGSJT","yahoo-finance-forex","Fetches real-time FOREX news and market data from Yahoo Finance for major currency pairs (EUR\u002FUSD, GBP\u002FUSD, USD\u002FJPY, etc.). Analyzes sentiment and provides trading context.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fyahoo-finance-forex\u002FSKILL.md",[283],{"categoryNo":284,"categoryName":285,"taxonomyType":236,"taxonomyTypeDesc":237},"01JBSEEDCAT000000000000040","财务",[287,288,289,290,291,292,293,294,295,296,297,298,299,300,301,302,303,304,305,306],"finance","yahoo finance","forex","yahoo","yahoo finance forex","github","https","sentiment","real-time","python3","install","Market","usd","news","time","currency","gbp","pip","market data","for",{"skillNo":308,"name":309,"nameZh":6,"description":310,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":311,"categories":312,"keywords":314,"publishedAt":43},"01M6Q5QM6VS84WZVM7BKXXPF32","yahoo-finance","This skill should be used when the user asks to \"get stock prices\", \"check stock quotes\", \"look up earnings\", \"get financial data\", \"find trending stocks\", or needs stock market data from Yahoo Finance.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fyahoo-finance-cli\u002FSKILL.md",[313],{"categoryNo":284,"categoryName":285,"taxonomyType":236,"taxonomyTypeDesc":237},[287,288,290,315,316,317,318,319,320,298,321,322,323,324,325,326,327,328,255,305],"cli","homebrew","find","check","brew","up","local","get","stock","link","trending","sf","quotes","ln"]