TomSzenessy/AI_TEMPLATE
Self-healing starting point for AI-agent-built projects of any size: docs bound to code, generated host adapters, delegation roles, and deterministic gates for any agent or model.
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 --transport http context7 https://mcp.context7.com/mcp -H 'CONTEXT7_API_KEY: YOUR_VALUE' $ claude mcp add --transport http exa https://mcp.exa.ai/mcp -H 'x-api-key: YOUR_VALUE' $ claude mcp add playwright -- npx -y @playwright/mcp@0.0.83 $ npx degit TomSzenessy/AI_TEMPLATE/.claude ./rig-ai_template # 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.
$ claude mcp add --transport http context7 https://mcp.context7.com/mcp -H 'CONTEXT7_API_KEY: YOUR_VALUE' $ claude mcp add --transport http exa https://mcp.exa.ai/mcp -H 'x-api-key: YOUR_VALUE' $ claude mcp add playwright -- npx -y @playwright/mcp@0.0.83
$ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/agent-handover .claude/skills/agent-handover $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/brand-sweep .claude/skills/brand-sweep $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/design-tokens .claude/skills/design-tokens $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/parity-check .claude/skills/parity-check $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/product-kickoff .claude/skills/product-kickoff $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/product-recon .claude/skills/product-recon $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/quality-loop .claude/skills/quality-loop $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/repository-audit .claude/skills/repository-audit $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/review-mining .claude/skills/review-mining $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/stack-foundation .claude/skills/stack-foundation $ npx degit TomSzenessy/AI_TEMPLATE/.agents/skills/ux-quality .claude/skills/ux-quality
$ curl -fsSL --create-dirs -o .claude/agents/critic.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/critic.md $ curl -fsSL --create-dirs -o .claude/agents/doc-gardener.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/doc-gardener.md $ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/researcher.md $ curl -fsSL --create-dirs -o .claude/agents/scout.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/scout.md $ curl -fsSL --create-dirs -o .claude/agents/skill-scout.md https://raw.githubusercontent.com/TomSzenessy/AI_TEMPLATE/main/.claude/agents/skill-scout.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 (3)
| server | source | est. tokens |
|---|---|---|
| Context7 | remote · remote | 1.2k |
| Exa | remote · remote | 1.5k |
| Playwright MCP · "playwright" | npm | 7.5k |
Skills (11)
agent-handoverbrand-sweepdesign-tokensparity-checkproduct-kickoffproduct-reconquality-looprepository-auditreview-miningstack-foundationux-quality
Subagents (6)
| critic model: opus | Fresh, read-only reviewer that tries to falsify a completion claim against the original acceptance criteria, the diff, and the real artifact. Use before declaring high-risk or subjective work done; th |
| doc-gardener model: haiku | Docs-only maintainer. Fixes stale or dead doc bindings, broken command references, and navigation drift reported by make garden or the stop gate, so documentation keeps matching what the code actually |
| implementer model: sonnet | Builds one bounded deliverable from a brief - smallest coherent change, tests first for bugs, owning docs updated in the same change, verified with the declared commands. Use for disjoint, well-specif |
| researcher model: sonnet | Answers an external-knowledge question (current library APIs, platform rules, prior art, error messages) from primary sources using web search, docs MCP servers, and a browser, returning cited, versio |
| scout model: haiku | Read-only locator. Use before reading many files yourself - finds where something lives, who owns it, how it is wired, and what broke there before, then returns a short map with file:line evidence. |
| skill-scout model: sonnet | Finds and vets a missing capability (skill, MCP server, or tool) without installing it - checks overlap with existing skills first, inspects candidates for provenance and risk, and returns a ready-to- |
Hooks (4)
| event | matcher | runs |
|---|---|---|
| SessionStart | * | "$CLAUDE_PROJECT_DIR/tools/repoctl" hook session-start |
| PreCompact | * | "$CLAUDE_PROJECT_DIR/tools/repoctl" hook pre-compact |
| PostToolUse | Edit|Write|MultiEdit|NotebookEdit | "$CLAUDE_PROJECT_DIR/tools/repoctl" hook after-edit |
| Stop | * | "$CLAUDE_PROJECT_DIR/tools/repoctl" hook stop |
Permissions
deny (0)
—
ask (0)
—
allow (18)
Bash(make start)
Bash(make next)
Bash(make ui-review)
Bash(make done)
Bash(make check)
Bash(make test)
Bash(make verify)
Bash(make map)
Bash(make where:*)
Bash(make garden)
Bash(make risk)
Bash(make capabilities)
Bash(make sync)
Bash(make handover)
Bash(make inventory)
Bash(make resources)
Bash(make similar:*)
Bash(make new:*)
Similar rigs
witekbobrowski/dotfiles
🗂 Automated custom macOS config
Orchestrator 988 tok ·
ifitsmanu/claude-template
The best Claude Code setup template. 14 commands, 5 agents, 6 MCP servers, auto-formatting hooks. Based on Boris Cherny's workflow.
Fort Knox 17.2k tok ·
codeverbojan/claude-code-kickstart
Production-grade agentic workflow template for Claude Code. 4-layer architecture, task playbooks, session memory, sub-agents, one-click install.
Orchestrator 10.2k tok ·
imMamdouhaboammar/agent-kernel
Local governance, memory, failure lessons, and architecture guardrails for Claude Code, Codex, Cursor, Gemini CLI, and other AI coding agents.
MCP Hoarder 34.4k tok ·