TheCaptainCat/squads
The coordination layer for a team of AI agents working on one - stable IDs, roles, and handoffs, all in plain markdown.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit TheCaptainCat/squads/.claude ./rig-squads # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit TheCaptainCat/squads/.claude/skills/launch-dev-host .claude/skills/launch-dev-host $ npx degit TheCaptainCat/squads/.claude/skills/releasing-squads .claude/skills/releasing-squads $ npx degit TheCaptainCat/squads/.claude/skills/sq-greeting .claude/skills/sq-greeting $ npx degit TheCaptainCat/squads/.claude/skills/sq-how-to .claude/skills/sq-how-to $ npx degit TheCaptainCat/squads/.claude/skills/sq-memory .claude/skills/sq-memory $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-architect .claude/skills/sq-role-architect $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-devops .claude/skills/sq-role-devops $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-manager .claude/skills/sq-role-manager $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-product-owner .claude/skills/sq-role-product-owner $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-python-dev .claude/skills/sq-role-python-dev $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-qa .claude/skills/sq-role-qa $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-reviewer .claude/skills/sq-role-reviewer $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-tech-lead .claude/skills/sq-role-tech-lead $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-tech-writer .claude/skills/sq-role-tech-writer $ npx degit TheCaptainCat/squads/.claude/skills/sq-role-typescript-dev .claude/skills/sq-role-typescript-dev
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/devops.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/devops.md $ curl -fsSL --create-dirs -o .claude/agents/manager.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/manager.md $ curl -fsSL --create-dirs -o .claude/agents/product-owner.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/product-owner.md $ curl -fsSL --create-dirs -o .claude/agents/python-dev.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/python-dev.md $ curl -fsSL --create-dirs -o .claude/agents/qa.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/qa.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/tech-lead.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/tech-lead.md $ curl -fsSL --create-dirs -o .claude/agents/tech-writer.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/tech-writer.md $ curl -fsSL --create-dirs -o .claude/agents/typescript-dev.md https://raw.githubusercontent.com/TheCaptainCat/squads/main/.claude/agents/typescript-dev.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (15)
launch-dev-hostreleasing-squadssq-greetingsq-how-tosq-memorysq-role-architectsq-role-devopssq-role-managersq-role-product-ownersq-role-python-devsq-role-qasq-role-reviewersq-role-tech-leadsq-role-tech-writersq-role-typescript-dev
Subagents (10)
| architect model: opus | System design and architecture decisions (ADRs). |
| devops model: sonnet | CI/CD, infrastructure, and releases. |
| manager model: opus | Default agent: triages the operator's request and routes it to the right specialist. |
| product-owner model: sonnet | Requirements, user stories, and backlog priorities. |
| python-dev model: sonnet | Implements Python code following the project's guides and standards. |
| qa model: sonnet | Designs and runs tests; verifies behaviour against acceptance criteria. |
| reviewer model: opus | Reviews code changes for correctness, clarity, and consistency. |
| tech-lead model: opus | Coordination and breaking features into tasks. |
| tech-writer model: sonnet | Documentation and guides. |
| typescript-dev model: sonnet | Implements Typescript code following the project's guides and standards. |
Permissions
deny (0)
—
ask (0)
—
allow (5)
Bash(sq:*)
Bash(squads:*)
Bash(uv run sq:*)
Skill(sq-greeting)
Skill(sq-how-to)
Similar rigs
xhulz/nina
Harness orchestration for Claude Code: composes a gated multi-agent pipeline into projects from versioned layers, holds its loop caps with hooks, measures it from transcripts, and turns recurring mistakes into rules.
Orchestrator 12.0k tok · · Q25
vannt-dev/governed-agent-sdlc
Ship with agents. Keep humans in control. Cross-platform governance toolkit for AI coding agents.
Fort Knox 632 tok · · Q24
eugeek/claude-config-tdd
Claude Code config with TDD workflow, plan -> tests -> red -> implementation -> green -> review -> shipment
Orchestrator 5.8k tok · · Q29
navox-labs/agents
15-agent AI engineering team for Claude Code. Full sprint cycle: Think, Plan, Build, Review, Test, Ship, Reflect. Zero dependencies.
Orchestrator 5.1k tok · · Q5