dispatching-parallel-agents
Identify independent failures and spawn concurrent agents. Turns sequential N×time into 1×time.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Identify independent failures and spawn concurrent agents. Turns sequential N×time into 1×time.
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
Deterministic AI development workflow orchestrator. Manages task boards (.jonggrang/jonggrang-tasks.json), runs 16-phase pipelines (BUGFIX/SMALL/MEDIUM/LARGE), installs lifecycle hooks for Claude Code and OpenCode, and coordinates five specialist roles (Lead/Developer/Reviewer/TestLead/Tester). Use when planning features, executing multi-step development workflows, or orchestrating AI coding agents. Run via npx jonggrang.
16-phase feature orchestration workflow. Runs in the main thread as Kernel Mode. Coordinates the Lead→Developer→Reviewer→TestLead→Tester assembly line.
Protocol for agents to write structured outputs that survive session resets and are findable by the orchestrator.
GitHub workflow automation with the gh CLI — pull requests, issues, releases, the REST/GraphQL API, and safe message escaping. Auth via GH_TOKEN (set in global settings, attached to the sandbox env).
GitLab workflow automation with the glab CLI — merge requests, issues, CI, releases, and safe message escaping. Auth via GITLAB_TOKEN (set in global settings, attached to the sandbox env).
Prevents premature exit and infinite loops. Three mechanisms: Completion Promises, Scratchpads, Loop Detection.
| name | dispatching-parallel-agents |
| description | Identify independent failures and spawn concurrent agents. Turns sequential N×time into 1×time. |
| type | orchestrate |
| tier | core |
| trigger | run in parallel, multiple independent tasks, concurrent agents, parallel execution |
When you have multiple independent failures or tasks that don't depend on each other, dispatch them in parallel. This skill defines how to identify, split, and recombine parallel work.
Tasks are independent if:
task-A modifies src/auth.ts
task-B modifies src/payments.ts
task-C modifies src/notifications.ts
→ All independent → dispatch in parallel
task-A creates User model
task-B creates Auth that imports User
→ task-B blocked_by task-A → serial only
Use the Task tool to spawn multiple agents simultaneously:
Spawn Task 1: {role: "developer", task: "implement auth module", files: ["src/auth.ts"]}
Spawn Task 2: {role: "developer", task: "implement payments module", files: ["src/payments.ts"]}
Spawn Task 3: {role: "tester", task: "fix auth-abort.test.ts failures", files: ["tests/auth-abort.test.ts"]}
These run concurrently. Time to resolution: 1× instead of 3×.
To prevent race conditions, each parallel agent registers its file ownership:
Lock file: .jonggrang/locks/{agentId}.lock
{
"agent_id": "developer-auth-001",
"role": "developer",
"files": ["src/auth.ts", "src/auth.test.ts"],
"acquired_at": "2024-01-01T00:00:00Z"
}
Before writing any file:
After completing task:
After all parallel agents complete:
Parallel test fixing:
Test failures: auth-abort.test.ts (3), batch.test.ts (2), race.test.ts (1)
→ Spawn 3 tester agents in parallel, one per file
Independent feature implementation:
Feature: user dashboard
Tasks: [create user profile component, create activity feed component, create settings panel]
→ All independent → dispatch in parallel
Multi-domain reviews:
Review needed: backend changes (3 files) + frontend changes (4 files)
→ Spawn backend-reviewer + frontend-reviewer in parallel