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/strategy-red-team-review.md
View on GitHub# Strategy: red-team-review (adversarial review loop)
Orchestrated adversarial review loop pairing iterative plan/code generation with independent critique personas (Architecture Skeptic, Security Auditor, TDD Contract Reviewer) until an "Approved" verdict is earned.
## Critical Constraints
1. **Fan-Out Trio Invariant**: Multi-persona reviews must dispatch to all 3 Graph Planning roles: Architecture Skeptic (`architect-review`), Security / Edge-Case Auditor (`security-auditor`), and TDD Contract Reviewer (`tdd-contract-reviewer`).
2. **Directory Isolation**: Save review packets and critique outputs to isolated directories (e.g. `.history/review-iteration-1/`) so baseline artifacts are never destructively overwritten.
3. **Convergence Limit**: If 2-3 rounds pass without approval, stop looping and escalate outstanding disagreements to human/orchestrator tie-breaker.
4. **Non-Interactive Input Redirection**: Always append `< /dev/null` to headless CLI commands to prevent `SIGTTIN` hangs.
## Quick start
```bash
# Bundle context with the bundling-context skill (dev-utils)
python3 plugins/dev-utils/skills/bundling-context/scripts/bundle.py --manifest manifest.json --bundle temp/review/bundle.md
```
## Workflow
1. **Research & Packet Generation**: Draft `red-team-prompt.md`, prepare `manifest.json`, and run `bundling-context` in Multi-Persona Fan-Out Mode to generate persona packets.
2. **Select Backend**: Interactively confirm CLI backend (`agy`, `claude`, `copilot`, `codex`) and model tier with user.
3. **Parallel Dispatch**: Dispatch review bundles in parallel to designated persona reviewers.
4. **Synthesize Feedback**: Capture verdicts. If "More Research Needed", iterate with targeted questions. If consensus is reached, advance.
5. **Handoff**: On approval, terminate review loop and pass approved artifacts back to parent orchestrator.
## Verification
```bash
# Execute project unit tests to verify no regressions were introduced
pytest tests/ -k "not test_health_check_signing_status"
```
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: 472k | 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 |