Zejnilovic/agd
Install, update and combine agentic setups from git repositories
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 Zejnilovic/agd/.claude ./rig-agd # 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 Zejnilovic/agd/.claude/skills/commits-and-prs .claude/skills/commits-and-prs $ npx degit Zejnilovic/agd/.claude/skills/context-md .claude/skills/context-md $ npx degit Zejnilovic/agd/.claude/skills/debugging .claude/skills/debugging $ npx degit Zejnilovic/agd/.claude/skills/diagrams .claude/skills/diagrams $ npx degit Zejnilovic/agd/.claude/skills/implementation .claude/skills/implementation $ npx degit Zejnilovic/agd/.claude/skills/project-profile .claude/skills/project-profile $ npx degit Zejnilovic/agd/.claude/skills/project-scan .claude/skills/project-scan $ npx degit Zejnilovic/agd/.claude/skills/session-retro .claude/skills/session-retro $ npx degit Zejnilovic/agd/.claude/skills/task-plans .claude/skills/task-plans $ npx degit Zejnilovic/agd/.claude/skills/tdd .claude/skills/tdd
$ curl -fsSL --create-dirs -o .claude/agents/analyst.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/analyst.md $ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/developer.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/developer.md $ curl -fsSL --create-dirs -o .claude/agents/documenter.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/documenter.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/security-reviewer.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/security-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/tester.md https://raw.githubusercontent.com/Zejnilovic/agd/main/.claude/agents/tester.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 (10)
commits-and-prscontext-mddebuggingdiagramsimplementationproject-profileproject-scansession-retrotask-planstdd
Subagents (7)
| analyst model: haiku | Discovers the lay of the land before planned work. Scans a project to record its purpose, tech stack, structure and commands, or turns a task's notes in .local_context into the discovery part of a pla |
| architect model: sonnet | Designs the approach for a change that spans modules, adds a component or dependency, or needs a design decision, and writes it into the task's plan file as steps, diagrams and decisions. Not for smal |
| developer model: haiku | Implements a well-defined task or plan steps test-first, in small verified increments. Use for work no framework developer covers (scripts, tools, libraries, infrastructure, other stacks) or small cha |
| documenter model: haiku | Keeps documentation true to the code, CONTEXT.md files first, then README and user-facing docs. Use when asked to update, check or write docs, to close a finished plan, or to add CONTEXT.md files to a |
| reviewer model: sonnet | Reviews a change for real problems (bugs, missing tests for important behavior, stale CONTEXT.md, over-engineering) without nitpicking, and checks that a plan's goal was actually met. Use before commi |
| security-reviewer model: sonnet | Reviews a change for exploitable security problems such as injection, broken access control, leaked secrets, weak crypto or risky dependencies, and reports only issues with a plausible attack path. Us |
| tester model: haiku | Finds untested behavior that matters in a change or module, writes the missing tests, and reports the bugs they expose. Also turns a bug report into a failing test. |
Similar rigs
Zejnilovic/agentic-defaults
Instructions, agents and skills for coding agents (Claude Code, GitHub Copilot). Test-first, cheap models, context that stays current. Install with agd.
Orchestrator 5.6k tok · · Q10
FCabiddu/pocket-it
Agent system for Claude Code: analyst, architect, task-writer, developer, reviewer, security and tester, with hooks and scripts that enforce the gates
Orchestrator 3.2k tok · · Q5
BillDX/spicy-to-dark-factory-starter
Workshop starter for 'From Spicy Autocomplete to Dark Factory: Agentic Coding in Practice' — a FastAPI mini-app with a full Claude Code harness (agents, slash commands, hooks, sandbox settings) and self-paced exercises.
Orchestrator 706 tok · · Q18
vannt-dev/aiws-factory
AIWS – AI Software Factory: turns a requirement into tested code through gated phases (analysis → design + test spec → human approval → per-task implementation with unit tests → review → PR). Language-agnostic, Claude Code first.
Fort Knox 1.5k tok · · Q19