speculator-orchestration
使用spec CLI进行多代理任务编排。指导代理声明任务、并行执行、阻塞和协调。使用多个代理、声明任务或管理并行工作时使用。
📖 中文导读:使用spec CLI进行多代理任务编排。指导代理声明任务、并行执行、阻塞和协调。使用多个代理、声明任务或管理并行工作时使用。 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。
name: speculator-orchestration description: Multi-agent task orchestration using the spec CLI. Guides agents on claiming tasks, parallel execution, blocking, and coordination. Use when working with multiple agents, claiming tasks, or managing parallel work.
Speculator Orchestration
Core Principle
Claim before you code.
Never start work on a task without atomically claiming it first. This prevents duplicate effort and merge conflicts in multi-agent scenarios.
Quick Reference
| Command | Purpose |
|---|---|
spec tasks ready |
Find tasks available for work (todo, unblocked, dependencies complete) |
spec task claim <id> |
Atomically claim a task (marks as in_progress) |
spec task complete <id> |
Mark a task as done |
spec task block <id> --reason <reason> |
Record a blocker with reason |
spec task unblock <id> |
Remove a block |
spec task release <id> |
Return a claimed task to todo |
spec task add --feature <id> --title <title> |
Create a new task or subtask |
spec task depend <id> --on <other-id> |
Add a dependency between tasks |
Workflows
- Claiming Tasks - How to find and claim work
- Completing Tasks - When and how to mark work done
- Blocking Tasks - When and how to record blockers
Patterns
- Parallelization - Which tasks can run concurrently
- Decomposition - Breaking large tasks into subtasks
Examples
- Multi-Agent Session - Two agents working concurrently
The Orchestration Loop
1. spec tasks ready → find available work
2. Evaluate conflicts → check for file overlap with other in-progress tasks
3. spec task claim <id> → atomically claim
4. Implement → do the work
5. spec task complete <id> | spec task block <id> --reason <reason> → finish or record blocker
6. Repeat
Important Notes
- Always check
spec tasks readybefore starting work - If claim fails, another agent got it first - move on
- Block tasks with clear, actionable reasons
- Release tasks if you can't complete them
- Create subtasks for large pieces of work