agentrigs
+ submit rig
~ / rigs / Trolzie / claude-loop-systems-demo

Trolzie/claude-loop-systems-demo

A small demo of Markdown-as-system AI dev loops with Claude Code: issues → agents → skills → verification.

↗ GitHub ★ 0 MIT updated 5mo ago project Claude Code
share on X
ARCHETYPE
Pragmatist
A balanced, no-drama setup: some rules, some tools, nothing extreme.
CONTEXT TAX · EVERY TURN
~480 tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
2/5
Pre-tool screening hook · No YOLO mode · details

Copy this rig

# review before running: this installs third-party code
$ npx degit Trolzie/claude-loop-systems-demo/.claude ./rig-claude-loop-systems-demo  # 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 Trolzie/claude-loop-systems-demo/.claude/skills/issue-loop .claude/skills/issue-loop
$ npx degit Trolzie/claude-loop-systems-demo/.claude/skills/verify .claude/skills/verify
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/Trolzie/claude-loop-systems-demo/main/.claude/agents/architect.md
$ curl -fsSL --create-dirs -o .claude/agents/commit.md https://raw.githubusercontent.com/Trolzie/claude-loop-systems-demo/main/.claude/agents/commit.md
$ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/Trolzie/claude-loop-systems-demo/main/.claude/agents/implementer.md
$ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/Trolzie/claude-loop-systems-demo/main/.claude/agents/reviewer.md

Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash\\(.*\\bgit\\b[^\\)]*\\b(?:commit|push)(?:\\b|:)",
        "hooks": [
          {
            "type": "command",
            "command": "npm run verify"
          }
        ]
      }
    ]
  }
}

Skills (2)

issue-loopverify

Subagents (4)

architect
model: sonnet
Turn a GitHub issue into a tiny, testable plan with explicit assumptions. Output should be actionable and bounded.
commit
model: sonnet
Create a commit on main that closes the GitHub issue (Fixes #N) and push it.
implementer
model: sonnet
Implement the approved plan with the smallest diff, update/add tests, then run verification.
reviewer
model: sonnet
Skeptical reviewer. Checks scope vs issue acceptance criteria, tests, and obvious edge cases. Must output APPROVED or CHANGES REQUIRED.

Hooks (1)

eventmatcherruns
PreToolUseBash\(.*\bgit\b[^\)]*\b(?:commit|push)(?:\b|:)npm run verify

Similar rigs

Saba-Var/claude-ward
A read-only tripwire that watches Claude Code's local config and reports suspicious changes.
★ 2 project Claude CodeCodex 2 agents 3mo ago
Pragmatist 1.8k tok ·
homeofe/improvements
Portable, LLM-agnostic framework for multi-model AI workflows: model routing, structured agent handoff (AAHP), and a phased agent pipeline. MIT.
★ 0 project Claude Code 6 agents 1 hooks 1mo ago
Orchestrator 1.4k tok ·
0xT4sty/claude-code-dotfiles
—
★ 0 setup Claude Code 3 agents 5mo ago
Pragmatist 979 tok ·
IntuitivePhella/AI-PhellOS
Portable AI development workflow template and CLI for spec-driven, test-aware software delivery with Codex, Claude Code, and other coding agents.
★ 2 project Claude CodeCodex 8 agents 4mo ago
Orchestrator 3.1k tok ·
copied ✓