Help us improve
Share bugs, ideas, or general feedback.
From Captain's Cabinet
When to use Agent Teams (Sonnet 4.6 subagents) vs orchestrator (Fable 5). TeamCreate patterns + model routing for parallel research, eval running, and adversarial review.
npx claudepluginhub nate-step/captains-cabinet --plugin captains-cabinetHow this skill is triggered — by the user, by Claude, or both
Slash command
/captains-cabinet:agent-team-workflowThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**Status:** promoted
Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Explores codebases via GitNexus: discover repos, query execution flows, trace processes, inspect symbol callers/callees, and review architecture.
Share bugs, ideas, or general feedback.
Status: promoted Created by: foundation Date: 2026-04-05 Validated against: feature implementation, bug fixes, quick patches Usage count: 0
When the CTO needs to make any code changes to the product codebase (/workspace/product/). All code changes go through Agent Teams -- never edit product code directly.
Use Agent Teams, NOT sub-agents. These are fundamentally different:
CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 is already set in the environment.
TeamCreate: "SEN-XXX: [feature description]"
- worker (Sonnet 4.6): "Implement [feature]. Files: [list]. Tests: [list]. When done, notify reviewer."
- reviewer (Sonnet 4.6): "Review worker's output. Check: correctness, tests pass, no regressions. Message worker with issues. Iterate until clean."
Worker implements -> reviewer reviews -> worker fixes -> reviewer approves -> you push/merge/deploy.
You only handle: reading specs, planning architecture, creating the team, then push -> CI -> merge -> deploy after the team is done.
Use a single Agent with isolation: worktree:
Agent tool:
- prompt: "Fix [description] in [file]. Run tests."
- model: sonnet
- isolation: worktree
isolation: worktree for code changesrecord-experience.sh with tag "crew"Code changes are implemented by Agent Teams with built-in review, preserving CTO context for architecture and deployment decisions.
Extracted from CTO role definition (/.claude/agents/cto.md) Agent Teams section. Moved to skill file per framework guidelines that procedural content belongs in skills, not role definitions.