菜单
Automated plan execution pipeline. Use when running /orchestrate on an /interactive-planning output. Handles plan ingestion, review, skill matching, tech research, user gating, and phased execution with worktree isolation, testing, code review, and incremental commits.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | orchestrator |
| description | Automated plan execution pipeline. Use when running /orchestrate on an /interactive-planning output. Handles plan ingestion, review, skill matching, tech research, user gating, and phased execution with worktree isolation, testing, code review, and incremental commits. |
| metadata | {"author":"shihwesley","version":"1.0.0"} |
| user-invocable | true |
| allowed-tools | ["Bash","Glob","Grep","Read","Write","Edit","Task","AskUserQuestion","TaskCreate","TaskUpdate","TaskList","Skill","ToolSearch","WebSearch"] |
Takes /interactive-planning output and executes it through a 6-stage pipeline: ingest, review, skill-match, research, user gate, and phased execution with worktree isolation.
/interactive-planning produces a plan directory with task_plan.md or manifest.md--resume)task_plan.md or manifest.md + specs), discover project docs (CLAUDE.md, codebase map)--resume)The orchestrator picks classic or team mode per-phase based on task count, shared files, and dependencies:
| Condition | Mode |
|---|---|
| 1-2 tasks | classic |
| 3+ tasks, all independent | classic |
| 4+ tasks with shared files or deps | team |
| Low/medium confidence, 3+ tasks | team |
--team flag | team (forced) |
--no-team flag | classic (forced) |
| Flag | Effect |
|---|---|
--dry-run | Run stages 1-4, show plan at stage 5, stop |
--resume | Skip completed phases, skip stage 2 review |
--max-parallel N | Max agents per batch (default: 3) |
--model sonnet|opus|haiku | Model for dispatched agents |
--phase N | Execute only phase N |
--team | Force Agent Teams mode |
--no-team | Force classic Task dispatch |
All component specifications live in references/:
| File | Role |
|---|---|
references/plan-ingester.md | Parse plan files, handle spec-driven and task-based modes |
references/plan-reviewer.md | Review and restructure plans (sizing, gaps, clarity) |
references/skill-matcher.md | 3-step skill resolution cascade |
references/tech-researcher.md | Fetch docs, produce cheat sheets, manage knowledge store |
references/skill-registry.json | Keyword-to-skill/agent mappings |
references/worktree-manager.md | Worktree lifecycle per phase |
references/agent-dispatcher.md | Dual-mode dispatch (team or classic) |
references/team-lead.md | Agent Teams lead coordination protocol |
references/phase-runner.md | Self-contained protocol for one phase |
references/phase-finisher.md | Test, review, commit chain (runs inside phase-runner) |
The main orchestrator stays lean (~800 tokens per phase) by delegating all heavy work to phase-runner subagents. Each phase-runner gets its own context window and returns a ~20-line structured result. A 5-phase plan uses roughly 4k tokens for execution.
# Standard execution
/orchestrate ./plans/my-feature
# Preview without executing
/orchestrate --dry-run ./plans/my-feature
# Resume after interruption
/orchestrate --resume ./plans/my-feature
# Execute specific phase with limited parallelism
/orchestrate --phase 2 --max-parallel 2 ./plans/my-feature
# Force team mode
/orchestrate --team ./plans/my-feature
PLAN_DIR to an absolute path via realpath. Make sure the directory exists and contains task_plan.md or manifest.md.general-purpose agent type with a warning. Add keywords to your plan's task descriptions to improve matching.--resume to skip stages on retry.File-based planning with interactive gates and native task tracking. Use when user says /plan, needs to break a complex feature into phases, or wants structured implementation planning with user approval at key decision points. Supports task mode (single plan file) and spec mode (multi-file architecture).
[What it does — one sentence]. Use when user [trigger phrases users would actually say]. Handles [key capabilities]. Do NOT use for [negative triggers if needed to prevent over-firing].
Create a navigable skill graph for a domain from research output or existing skills/docs. Use when user says /create-skill-graph, wants to organize skills into MOCs, build a knowledge graph, or structure domain docs into a navigable hierarchy. Accepts a domain slug with --from-research or --from-existing flags.
Release a shihwesley plugin with semver versioning, conventional commit parsing, changelog generation, and GitHub Release creation. Use when user says /release, wants to tag a new version, publish a release, bump version, or create a changelog. Supports patch/minor/major and --dry-run.
Navigation protocol template for skill graphs. Copy and adapt per domain to teach agents how to efficiently traverse MOC hierarchies. Agent-only — not user-invocable. Use when building a new skill graph that needs a traversal protocol.