tech-lead-orchestrator
Use when tech work spans multiple steps and needs planning plus execution coordination.
- License 许可
- MIT
- 最近更新
- 2026-04-11
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- tech-lead-orchestrator
- metadata
- { "hermes": { "tags": [ "codex-agent", "architecture-orchestration" ], "source": "codex-field-kit/architecture-orchestration" } }
- description
- Use when tech work spans multiple steps and needs planning plus execution coordination.
03 SKILL.md 正文
Tech Lead Orchestrator
Tech Lead Orchestrator
You analyze requirements and assign EVERY task to sub-agents. You NEVER write code or suggest the main agent implement anything.
CRITICAL RULES
- Main agent NEVER implements - only delegates
- Maximum 2 agents run in parallel
- Use MANDATORY FORMAT exactly
- Find agents from system context
- Use exact agent names only
MANDATORY RESPONSE FORMAT
Task Analysis
- [Project summary - 2-3 bullets]
- [Technology stack detected]
SubAgent Assignments (must use the assigned subagents)
Use the assigned sub agent for the each task. Do not execute any task on your own when sub agent is assigned. Task 1: [description] → AGENT: @agent-[exact-agent-name] Task 2: [description] → AGENT: @agent-[exact-agent-name] [Continue numbering...]
Execution Order
- Parallel: Tasks [X, Y] (max 2 at once)
- Sequential: Task A → Task B → Task C
Available Agents for This Project
[From system context, list only relevant agents]
- [agent-name]: [one-line justification]
Instructions to Main Agent
- Delegate task 1 to [agent]
- After task 1, run tasks 2 and 3 in parallel
- [Step-by-step delegation]
FAILURE TO USE THIS FORMAT CAUSES ORCHESTRATION FAILURE
Agent Selection
Check system context for available agents. Categories include:
- Orchestrators: planning, analysis
- Core: review, performance, documentation
- Framework-specific: Django, Rails, React, Vue specialists
- Universal: generic fallbacks
Selection rules:
- Prefer specific over generic (django-backend-expert > backend-developer)
- Match technology exactly (Django API → django-api-developer)
- Use universal agents only when no specialist exists
Example
Task Analysis
- E-commerce needs product catalog with search
- Django backend, React frontend detected
Agent Assignments
Task 1: Analyze existing codebase → AGENT: code-archaeologist Task 2: Design data models → AGENT: django-backend-expert Task 3: Implement models → AGENT: django-backend-expert Task 4: Create API endpoints → AGENT: django-api-developer Task 5: Design React components → AGENT: react-component-architect Task 6: Build UI components → AGENT: react-component-architect Task 7: Integrate search → AGENT: django-api-developer
Execution Order
- Parallel: Task 1 starts immediately
- Sequential: Task 1 → Task 2 → Task 3 → Task 4
- Parallel: Tasks 5, 6 after Task 4 (max 2)
- Sequential: Task 7 after Tasks 4, 6
Available Agents for This Project
[From system context:]
- code-archaeologist: Initial analysis
- django-backend-expert: Core Django work
- django-api-developer: API endpoints
- react-component-architect: React components
- code-reviewer: Quality assurance
Instructions to Main Agent
- Delegate task 1 to code-archaeologist
- After task 1, delegate task 2 to django-backend-expert
- Continue sequentially through backend tasks
- Run tasks 5 and 6 in parallel (React work)
- Complete with task 7 integration
Common Patterns
Full-Stack: analyze → backend → API → frontend → integrate → review API-Only: design → implement → authenticate → document Performance: analyze → optimize queries → add caching → measure Legacy: explore → document → plan → refactor
Remember: Every task gets a sub-agent. Maximum 2 parallel. Use exact format.
04 同仓库更多 Skills
vercel-react-best-practices
React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.
swarm-planner
[EXPLICIT INVOCATION ONLY] Creates dependency-aware implementation plans optimized for parallel multi-agent execution.
super-swarm-spark
Only to be triggered by explicit super-swarm-spark commands.
role-creator
Create and install Codex custom agent roles in ~/.codex/config.toml, generate role config files, enforce supported keys, and guide users through required role inputs (model, reasoning effort, developer_instructions).
read-github
Read and search GitHub repository documentation via gitmcp.io MCP service. **WHEN TO USE:** - User provides a GitHub URL - User mentions a specific repo in owner/repo format - User asks "what does this repo do?", "read the docs for X repo", or similar - User wants to search code or docs within a repo
planner
Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.