Skip to content

borghei/Claude-Skills / project-management/execution/agents-in-the-team

agents-in-the-team

Unverified

Run delivery when AI coding and ops agents take tickets. Use when setting agent delegation policy, writing agent-ready tickets, planning review capacity, measuring agent vs human delivery, or rolling agents out safely.

assets/rollout_plan_template.md

View on GitHub
# Agent Rollout Plan: <team name>

| Phase | Duration | Scope | Exit criteria to advance | Rollback trigger |
|-------|----------|-------|--------------------------|------------------|
| 0. Baseline | 2-4 weeks | No agents. Measure DORA five + review time with `agent_delivery_metrics.py` (all human) | Baseline recorded | - |
| 1. Pilot | 1-2 sprints | 1 team, agent-eligible tickets only (tests, docs, dependency patches), 2 named reviewers | Agent PR acceptance and post-merge rework within agreed gap of human baseline; no security finding | Any escaped defect with customer impact from an agent PR; any secret exposure |
| 2. Expand types | 2-3 sprints | Add low/medium-risk bugs and small features with good tests | Review capacity gate passing every sprint; change fail rate of deployments with agent changes within agreed gap | Change fail rate gap above tolerance two sprints running |
| 3. Expand teams | quarter | Second and third team adopt the policy | Each team has its own policy, reviewers and baseline | Team reports review overload in retro two sprints running |
| 4. Steady state | ongoing | Quarterly policy review against metrics | - | - |

## Decisions to record before Phase 1
- [ ] Which agent(s) and which trackers/repos they may access
- [ ] Agent identity and permission scopes (least privilege)
- [ ] Branch protection and required human approvals
- [ ] Hard-block list signed off by security
- [ ] Tolerance for the metrics gate (`--tolerance`) and who acts on a breach
- [ ] How agent-authored work is labelled for metrics and Sprint Review

## Communication
- Kickoff: why, what changes for reviewers, what does not change (accountability)
- Weekly during pilot: metrics snapshot + one learning
- Retro item every sprint: "What did the agents make harder?"
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: 472kUpdated Claude CodeCodexVerifiedInstall
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: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
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 availableStars: 323kUpdated Claude CodeCodexVerifiedInstall