원클릭으로
dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work. In this repository, package or lane branches must close through package-first evidence and integration handoff rather than the generic local merge or PR menu.
Use when executing repo-wide or lane-scoped SDK changes in this repository that must follow bounded change packages, qiniu-ai worktree/package commands, evidence bundles, review packets, promotion decisions, and integration-branch handoff. Trigger on requests to open a phase package, spawn or integrate a lane, prepare a review handoff, or land multi-subsystem SDK work without widening scope.
Four-phase debugging framework that ensures root cause investigation before attempting fixes. Never jump to solutions.
Use when starting feature work that needs isolation from the current workspace or before executing implementation plans. In this repository, prefer qiniu-ai worktree commands for lane or package-first delivery; fall back to generic git worktree flows only for non-package branches.
Run verification commands and confirm output before claiming success
Use when you have a spec or requirements for a multi-step task, before touching code. In this repository, multi-subsystem or lane-scoped work must create the package brief first and keep the plan aligned with the bounded package contract.
| name | dispatching-parallel-agents |
| description | Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies |
When you have multiple unrelated problems (different test files, different subsystems, different bugs), investigating them as one giant task wastes mental energy. Break them into focused, independent tasks and execute sequentially.
Core principle: One focused task at a time. Complete isolation between problem domains.
Use when:
Don't use when:
Group failures/tasks by what's broken:
Each domain is independent - fixing auth doesn't affect UI tests.
Rank by:
For each domain:
## Domain: [Name]
### Context
[Full context for this problem only]
### Goal
[Specific, measurable outcome]
### Constraints
- Don't touch code outside this domain
- Complete this before moving to next
### Execute
[Work through the problem]
### Summary
[What was found, what was fixed]
---
[Move to next domain]
After all domains complete:
Each task should be:
✅ Focused - One clear problem domain
Fix the 3 failing tests in auth.test.ts
✅ Self-contained - All context included
Errors:
1. "should validate token" - expects valid, got expired
2. "should refresh token" - timeout after 5000ms
3. "should logout" - session not cleared
✅ Specific about outcome
Outcome: All 3 tests passing, no new failures introduced
✅ Constrained
Constraint: Only modify auth/ directory
❌ Too broad: "Fix all the tests" ✅ Specific: "Fix auth.test.ts failures"
❌ No context: "Fix the race condition" ✅ Context: Include error messages, test names, file paths
❌ No constraints: Might refactor everything ✅ Constraints: "Only modify auth/ directory"
❌ Vague outcome: "Fix it" ✅ Specific: "All tests green, document root cause"
Scenario: 6 test failures across 3 files
Step 1: Identify Domains
- auth.test.ts: 3 failures (authentication)
- data.test.ts: 2 failures (data validation)
- ui.test.ts: 1 failure (rendering)
Step 2: Prioritize
Step 3: Execute Domain 1
## Domain: Authentication (auth.test.ts)
### Context
3 failures related to token handling...
### Execute
[Full focus on auth only]
[Found: Token expiry not checked correctly]
[Fixed: Added expiry validation]
### Summary
Root cause: Missing expiry check
Fix: Added validation in validateToken()
Tests: All 3 passing
Step 4: Execute Domain 2
## Domain: Data Validation (data.test.ts)
...
Step 5: Execute Domain 3
## Domain: UI Rendering (ui.test.ts)
...
Step 6: Integrate
## Integration Check
- All domains complete
- No conflicts between fixes
- Full test suite: 127 passing, 0 failing
- Ready to commit
Related failures: Investigate together first Need full context: Understanding requires entire system Exploratory: You don't know what's broken yet