ToruAI/toru-claude-agents
Multi-agent team for Claude Code: 7 AI specialists (strategy, dev, security, sales, design, data, research) that collaborate and challenge you. Dev-cycle workflow with machine-verifiable completion.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ claude mcp add megg -- npx -y megg@1.1.0 mcp $ npx degit ToruAI/toru-claude-agents/agents ./rig-toru-claude-agents/agents $ npx degit ToruAI/toru-claude-agents/skills ./rig-toru-claude-agents/skills $ npx degit ToruAI/toru-claude-agents/hooks ./rig-toru-claude-agents/hooks
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.
$ claude mcp add megg -- npx -y megg@1.1.0 mcp $ npx degit ToruAI/toru-claude-agents/skills/branch-workflow .claude/skills/branch-workflow $ npx degit ToruAI/toru-claude-agents/skills/cc-docs .claude/skills/cc-docs $ npx degit ToruAI/toru-claude-agents/skills/claude-code-mastery .claude/skills/claude-code-mastery $ npx degit ToruAI/toru-claude-agents/skills/dev-cycle .claude/skills/dev-cycle $ npx degit ToruAI/toru-claude-agents/skills/dev-finish .claude/skills/dev-finish $ npx degit ToruAI/toru-claude-agents/skills/dev-full-auto .claude/skills/dev-full-auto $ npx degit ToruAI/toru-claude-agents/skills/dev-qa .claude/skills/dev-qa $ npx degit ToruAI/toru-claude-agents/skills/dev-rc .claude/skills/dev-rc $ npx degit ToruAI/toru-claude-agents/skills/dev-security .claude/skills/dev-security $ npx degit ToruAI/toru-claude-agents/skills/megg-learn .claude/skills/megg-learn $ npx degit ToruAI/toru-claude-agents/skills/megg-state .claude/skills/megg-state $ npx degit ToruAI/toru-claude-agents/skills/openspec-workflow .claude/skills/openspec-workflow $ npx degit ToruAI/toru-claude-agents/skills/scout .claude/skills/scout $ npx degit ToruAI/toru-claude-agents/skills/self-improvement .claude/skills/self-improvement
$ curl -fsSL --create-dirs -o .claude/agents/arlo.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/arlo.md $ curl -fsSL --create-dirs -o .claude/agents/bob.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/bob.md $ curl -fsSL --create-dirs -o .claude/agents/garry.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/garry.md $ curl -fsSL --create-dirs -o .claude/agents/luna.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/luna.md $ curl -fsSL --create-dirs -o .claude/agents/maya.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/maya.md $ curl -fsSL --create-dirs -o .claude/agents/scout.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/scout.md $ curl -fsSL --create-dirs -o .claude/agents/sentinel.md https://raw.githubusercontent.com/ToruAI/toru-claude-agents/main/agents/sentinel.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} MCP servers (1)
| server | source | est. tokens |
|---|---|---|
| megg | npm | 2.5k |
Skills (14)
branch-workflowcc-docsclaude-code-masterydev-cycledev-finishdev-full-autodev-qadev-rcdev-securitymegg-learnmegg-stateopenspec-workflowscoutself-improvement
Subagents (7)
| arlo model: inherit | Finds the pattern in messy data and says which parts the numbers actually support. Use for analysis, metrics, validating an assumption before you build on it. |
| bob model: sonnet | Technical Builder - builds elegant systems, ensures feasibility, executes with quality. Use when code needs writing, systems need building, or technical problems need solving. |
| garry model: opus | Chief of Staff & Strategic Director. Use for strategic decisions, delegating to other agents, coordinating work across projects, vision alignment, and improving AI workflows. Use proactively when plan |
| luna model: inherit | Designs interfaces and flows that remove decisions instead of adding controls. Use for UI, UX flows, design systems, and accessibility. |
| maya model: inherit | Writes and pressure-tests client-facing communication — pitches, proposals, hard emails. Use when the audience is outside your team. |
| scout model: haiku | Codebase Explorer & Research Synthesizer. Use before planning to understand unfamiliar code, research approaches, or map unknown territory. Fast, thorough, read-only. |
| sentinel model: opus | Security specialist for code audits, vulnerability scanning, dependency checks, and security best practices. Use proactively after code changes or when implementing auth, crypto, or handling sensitive |
Similar rigs
leavemagic-cyber/earned-confidence
A behavior contract for AI coding agents — every rule added after a real failure. Ships with a verifier agent that cannot edit what it judges.
Pragmatist 3.6k tok ·
OneWave-AI/claude-skills
200+ production-ready Claude Code skills for sales, marketing, design, engineering, and AI agent architecture. Built and maintained by OneWave AI.
Minimalist 204 tok ·
uncoalesced/ArgusCouncil
Provenance locked, three agent research harness for any AI coding agent. A Scout downloads sources, an Archivist hashes and quotes them, an Integrator cites them into your paper, report, or dataset. A stdlib Python engine verifies every quo
Minimalist 1.2k tok ·
frontier-infra/adl
Agent Discipline Layer (ADL) — keep a coding agent honest and make it prove its work: layered guardrails + a /goal contract Warden verifies against reality and signs. Claude Code today; Grok/Codex CLI/Pi planned. (formerly Claude Layers)
Pragmatist 1.2k tok ·