| name | qa-lead-routing |
| description | Coordinates QA workflow across planning, writing, and execution agents. Use when user requests testing, quality assurance, or test documentation. |
| scope | core |
| user-invocable | false |
| context | fork |
QA Lead Routing Skill
Purpose
Coordinates QA team activities by routing tasks to qa-planner, qa-writer, and qa-engineer agents. This skill contains the coordination logic for orchestrating the complete quality assurance workflow.
QA Team Agents
| Agent | Role | Output |
|---|
| qa-planner | Test planning | QA plans, test scenarios, acceptance criteria |
| qa-writer | Documentation | Test cases, test reports, templates |
| qa-engineer | Execution | Test results, defect reports, coverage reports |
Routing Decision (Priority Order)
Before routing via Agent tool, evaluate Agent Teams eligibility first:
Self-check: Does this task need 3+ agents, shared state, or inter-agent communication? If yes, prefer Agent Teams over Agent tool. See R018 for the full decision matrix.
| Scenario | Preferred |
|---|
| Single QA phase (plan/write/execute) | Agent Tool |
| Full QA cycle (plan + write + execute + report) | Agent Teams |
| Quality analysis (parallel strategy + results) | Agent Teams |
| Quick test validation | Agent Tool |
Command Routing
QA Request → Routing → QA Agent(s)
test_planning → qa-planner
test_documentation → qa-writer
test_execution → qa-engineer
quality_analysis → qa-planner + qa-engineer (parallel)
full_qa_cycle → all agents (sequential)
Claude compatibility Agent calls only (R010 “Delegated Permission Ownership”): Pass mode: "bypassPermissions" when the active Claude session uses bypass permissions. Native Codex spawn_agent has no mode parameter; use the installed agent_type and active Codex runtime permissions instead.
Ontology-RAG Enrichment (R019)
If get_agent_for_task MCP tool is available, call it with the original query and inject suggested_skills into the agent prompt. Skip silently on failure.
Step 5: Soul Injection (R006)
If the selected agent has soul: true in frontmatter, read and prepend .codex/agents/souls/{agent-name}.soul.md content to the prompt. Skip silently if file doesn't exist.
Sequential Workflow Ordering
Full QA cycle follows sequential phases (each depends on the previous):
qa-planner → qa-writer → qa-engineer → qa-writer
(plan) (document) (execute) (report)
Parallel execution only for independent analyses (e.g., multi-module testing). See R009.
Sub-agent Model Selection
All QA agents use frontier/medium by default for balanced quality output.
No Match Fallback
When a QA task involves unfamiliar testing patterns or tools:
User Input → QA task with unrecognized tool/pattern
↓
Detect: Testing framework or QA methodology keyword
↓
Delegate to mgr-creator with context:
domain: detected QA tool/methodology
type: qa-engineer
keywords: extracted testing terms
skills: auto-discover from .codex/skills/
guides: auto-discover from templates/guides/
Examples of dynamic creation triggers:
- New testing frameworks (e.g., "Cypress E2E 테스트 작성해줘", "k6 부하 테스트 설계해줘")
- Specialized QA methodologies (e.g., "뮤테이션 테스트 전략 만들어줘")
- Performance/security testing tools not covered by existing agents
Sensitive-Path Delegation
Sensitive-path compatibility note: if this skill delegates work that touches .claude/**, .claude/outputs/**, templates/.claude/**, or read-only measurements of those paths, keep .codex/** edits on the normal Codex path. On Claude Code v2.1.121+ with bypassPermissions, direct writes to .claude/skills/, .claude/agents/, and .claude/commands/ are allowed; on v2.1.126+ that extends to broader protected paths. Only use /tmp/{skill}-{timestamp}.md as a legacy fallback when the target runtime is older or still prompts.
Usage
This skill is NOT user-invocable. It should be automatically triggered when the main conversation detects QA intent.
Detection criteria:
- User requests testing
- User mentions quality assurance
- User asks for test plan/cases/execution
- User requests QA metrics/reports
- System detects need for quality verification
