richfrem/agent-plugins-skills / plugins/agent-orchestration/.apm/skills/orchestrating-agent-work
orchestrating-agent-work
VerifiedInteractive orchestrator for complex engineering work. Analyzes the task, conducts a diagnostic interview to select the appropriate execution strategy (Directed Acyclic Graph versus Loop, and which Loop type), plans the artifact, and supervises execution and verification. Use for "how should I orchestrate this", "plan this workflow", "should this be a DAG or a loop", "run these jobs in parallel", "delegate this to a worker agent", or "run/resume this graph manifest". Not for single trivial edits or skill self-improvement.
references/routing-and-closure.md
View on GitHub# Routing, delegation and closure
Assesses incoming tasks, routes to specialized execution patterns, and enforces verification and closure.
## Constraints
- **Trust but verify**: Never blindly trust sub-agent outputs; mechanically run automated tests.
- **Inner loop isolation**: Inner Loop agents are strictly forbidden from running `git` commands.
- **Background execution**: Always redirect stdin (`< /dev/null`) when spawning sub-agents.
- **Zero autonomous merge**: Merging and pushing are strictly caller-managed. The orchestrator halts after verification and retrospective.
## Quick start
```bash
# Generate strategy packet for delegation
python3 scripts/agent_orchestrator.py packet --wp 001 --spec-dir handoffs
```
## Routing Decision Table
| Signal | Pattern | Follow |
|---|---|---|
| Security review, architecture decision, high risk | Adversarial Critique | [strategy-red-team-review.md](strategy-red-team-review.md) |
| Bounded code implementation, bug fix | Supervisor / Worker | [strategy-dual-loop.md](strategy-dual-loop.md) |
| Bulk migration, 10+ independent partitioned jobs | Parallel Swarm | [strategy-agent-swarm.md](strategy-agent-swarm.md) |
| Structural dependencies, parallel fan-out joined at barriers | Graph | [graph-planning.md](graph-planning.md), then [graph-execution.md](graph-execution.md) |
| GUI computer-use, telephony, external SaaS/commerce, deep research | Asynchronous Bot Handoff | [bot-delegation-patterns.md](bot-delegation-patterns.md) |
| Ambiguous routing, topology comparison | Decision Tree | strategy selection ([strategy-selection.md](strategy-selection.md)) |
## Workflow
1. **Assess & Route**: Match incoming signal against the Routing Decision Table.
2. **Package**: Author Task Packet via `agent_orchestrator.py packet`. For graph workflows, mandate compilation of `graph-manifest.json` via graph planning ([graph-planning.md](graph-planning.md)).
3. **Dispatch**: Dispatch inner worker into isolated worktree with stdin redirection (`< /dev/null`).
4. **Supervised Verify**: Audit diff and run tests via `agent_orchestrator.py verify`.
5. **Close & Retro**: Run `agent_orchestrator.py retro`. Stop after recording retrospective. Caller performs review, merge, and deployment.
## Verification
```bash
# Verify worker execution results
python3 scripts/agent_orchestrator.py verify --packet handoffs/task_packet_001.md --worktree .
# Ensure tests pass
pytest
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |