a-gusman-claude
Health Warn
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 8 GitHub stars
Code Pass
- Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
This framework transforms Claude Code into an autonomous development platform. It provides a suite of 52 internal agents and pre-configured "machines" to automate complex pipelines like building features, fixing bugs, and running tests.
Security Assessment
Overall risk: Medium. The system relies heavily on autonomous execution, shell commands, and file system access. The automated installation script fetches remote code via `curl | bash`, which requires careful manual review. Additionally, the auto-update feature silently executes `git pull` in the background, creating a potential supply chain risk if the upstream repository is ever compromised. While the light code scan found no dangerous patterns, hardcoded secrets, or excessive permission requests in the audited files, the tool's fundamental nature is to execute code and modify your local environment.
Quality Assessment
The project is highly active and documented under a safe MIT license. It is very new and currently has low community visibility, demonstrated by only 8 GitHub stars despite significant activity (464 commits). This low adoption means fewer external eyes have reviewed the codebase. However, the documentation is thorough, detailing exactly what the script installs, how the auto-updater works, and how to completely uninstall it.
Verdict
Use with caution — the underlying framework is actively maintained and transparent, but its autonomous nature, silent background updates, and low community oversight require you to inspect the installation script before running it.
Claude Code agent system — 46 agents, 60 skills, 31 rules, 7 hooks, 12 playbooks. 193+ sessions, 464 commits. Self-improving.
Gusman Claude Agent System
Transform Claude Code from a passive assistant into an autonomous development platform.
13 machines, 52 internal agents, 66 skills. Battle-tested across 193+ real sessions. Self-improving.
Why This Exists
Claude Code is powerful but generic. This framework adds:
- 13 numbered machines — autonomous pipelines with convergence loops (MERIDIAN pattern)
- Defense-in-depth safety with 4 layers of protection (hooks, rules, permissions, governance)
- Self-improvement pipeline that learns from failures and improves agent prompts
- Model routing that uses the right model for the right task (haiku for scans, opus for debugging)
- Parallel execution via Agent Teams for bug sprints, multi-module builds, and audits
- QAT (Quality Acceptance Testing) — AI-as-Judge validates output quality, not just functionality
- Issue pipeline — GitHub Issue → SPEC → Build → Verify → Test → PR (fully automated)
Quick Start
# One-line install (clones repo + symlinks into .claude/ + auto-update hook)
curl -fsSL https://raw.githubusercontent.com/andregusman-raiz/a-gusman-claude/main/install.sh | bash
# Or clone manually
git clone https://github.com/andregusman-raiz/a-gusman-claude.git ~/.gusman-claude
bash ~/.gusman-claude/install.sh
# Start Claude Code and use any machine
claude
# Then type: /ag-0-orquestrador [describe what you want]
# Or directly: /ag-1-construir implement user auth
What install.sh does
- Clones the repo to
~/.gusman-claude/(the source of truth) - Symlinks
agents/,skills/,rules/,hooks/,shared/into your.claude/ - Installs auto-update hook that
git pulls latest on every session start (~1h throttle)
Your files are never touched: settings.local.json, projects/, memory — all yours.
Auto-update
Every time you start Claude Code, a SessionStart hook checks for updates:
- Runs
git pull --ff-onlyin background (never blocks startup) - Throttled to max once per hour
- To disable:
bash install.sh --no-auto-updateor delete.claude/hooks/auto-update.sh - To update manually:
cd ~/.gusman-claude && git pull
Uninstall
bash ~/.gusman-claude/install.sh --uninstall
# Removes symlinks and repo. Your settings/memory preserved.
Architecture — 13 Machines
Each machine is an autonomous pipeline following the MERIDIAN pattern:
phases → convergence loop → state persistence → self-healing → artifacts.
The user interacts with 13 commands. Each machine internally orchestrates multiple agents.
ag-0 ORQUESTRADOR ← entry point (classifies intent → delegates to machine)
│
├── ag-1 CONSTRUIR feature, issue, refactor, optimize, ui, integrate
├── ag-2 CORRIGIR bugs, TypeScript errors, tech debt
├── ag-3 ENTREGAR preview, production, rollback
├── ag-4 TESTE-FINAL QAT, UX-QAT, benchmark, E2E, test-fix-retest
├── ag-5 DOCUMENTOS project docs, Office, organize, spelling
├── ag-6 INICIAR new project, setup, explore, research
├── ag-7 QUALIDADE MERIDIAN (5D autonomous QA, MQS ≥ 85)
├── ag-8 SEGURANCA SENTINEL (6D security+load+LGPD, SSS ≥ 80)
├── ag-9 AUDITAR FORTRESS (full audit: 5 machines sequentially)
├── ag-10 BENCHMARK Crawl SaaS platforms, screenshot, AI analysis, SPEC
├── ag-11 DESENHAR UI/UX design intelligence (67 styles, 96 palettes)
└── ag-12 SQL-TOTVS SQL Server (TOTVS RM) + PostgreSQL optimization
Safety Layers:
[PreToolUse Hooks] → [Rules (27)] → [Permissions] → [CLAUDE.md]
How Machines Work (MERIDIAN Pattern)
Every machine follows the same pattern:
- ASSESS — auto-detect sub-mode from user input
- Phase sequence — SPEC → PLAN → BUILD → VERIFY (varies by machine)
- Convergence — VERIFY loops back to BUILD if threshold not met (max 2-3 cycles)
- State persistence —
*-state.jsonallows--resumeafter interruption - Self-healing — failures trigger alternatives, never block
- Artifacts — tangible outputs (PR, SPEC, certificate, report)
Common Workflows
Build a Feature
/ag-1-construir implement user authentication with Clerk
→ ASSESS (feature mode) → SPEC → PLAN → BUILD → VERIFY → REVIEW → PR
Fix Bugs
/ag-2-corrigir login not working after Clerk update
→ ASSESS (single bug) → DIAGNOSE → FIX → VERIFY (loop until green) → PR
Fix Multiple Bugs
/ag-2-corrigir lista: [bug1, bug2, bug3, bug4, bug5]
→ auto-routes to batch mode → sprints with incremental commits → PR
Resolve GitHub Issue
/ag-1-construir issue #42
→ Fetch issue → SPEC → PLAN → BUILD → VERIFY vs SPEC → TEST → PR (closes #42)
Deploy to Production
/ag-3-entregar producao
→ PREFLIGHT (credentials + typecheck + lint + test + build) → DEPLOY → SMOKE → MONITOR
Quality Audit
/ag-7-qualidade https://myapp.vercel.app
→ MERIDIAN: SCOUT → SIEGE (5D) → FORGE (fix) → CONVERGE (MQS ≥ 85) → CERTIFICATE
Full Software Audit
/ag-9-auditar https://myapp.vercel.app
→ FORTRESS: MERIDIAN → SENTINEL → ARCHITECT → CONDUCTOR → LIGHTHOUSE → Fortress Score
Design UI/UX
/ag-11-ux-ui landing page minimalist dark mode
→ Design system selection, palette, typography, components, implementation
New Project from Scratch
/ag-6-iniciar projeto SaaS with Next.js + Supabase + Clerk
→ SCAFFOLD → SETUP → EXPLORE → ready for /ag-1-construir
Or Just Let the Orchestrator Decide
/ag-0-orquestrador [describe anything you want]
→ classifies into 1 of 13 machines → delegates automatically
Machine Details
| # | Command | Sub-modes | Internal Agents |
|---|---|---|---|
| 0 | /ag-0-orquestrador |
auto-route | — |
| 1 | /ag-1-construir |
feature, issue, refactor, optimize, ui, integrate, --validado | ag-especificar-solucao, ag-planejar-execucao, ag-implementar-codigo, ag-validar-execucao, ag-testar-codigo, ag-revisar-codigo |
| 2 | /ag-2-corrigir |
bug, tipos, batch, debt, triage | ag-depurar-erro, ag-corrigir-bugs, ag-corrigir-tipos |
| 3 | /ag-3-entregar |
preview, producao, rollback | ag-versionar-codigo, ag-pipeline-deploy, ag-smoke-vercel, ag-monitorar-producao |
| 4 | /ag-4-teste-final |
qat, ux-qat, benchmark, ciclo, e2e | ag-testar-qualidade-qat, ag-testar-ux-qualidade, ag-benchmark-qualidade, ag-ciclo-testes |
| 5 | /ag-5-documentos |
projeto, office, organizar, ortografia | ag-documentar-projeto, ag-gerar-documentos, ag-organizar-arquivos, ag-revisar-ortografia |
| 6 | /ag-6-iniciar |
projeto, ambiente, explorar, pesquisar | ag-criar-projeto, ag-preparar-ambiente, ag-explorar-codigo, ag-pesquisar-referencia |
| 7 | /ag-7-qualidade |
— | ag-meridian (MERIDIAN engine) |
| 8 | /ag-8-seguranca |
— | ag-sentinel (SENTINEL engine) |
| 9 | /ag-9-auditar |
— | ag-fortress (FORTRESS engine) |
| 10 | /ag-10-benchmark-software |
— | Playwright crawl + AI analysis |
| 11 | /ag-11-ux-ui |
plan, build, review | 67 styles, 96 palettes, 13 stacks |
| 12 | /ag-12-sql-totvs |
— | SQL Server + PostgreSQL patterns |
Plugins (Shortcuts — No Pipeline)
Plugins complement machines for quick, single-action tasks:
| Command | What |
|---|---|
/code-review |
Quick PR review (< 10 files) |
/commit |
Git commit (no branch-guard) |
/commit-push-pr |
Commit + push + PR |
/deploy |
Quick Vercel deploy |
/feature-dev |
Self-contained feature (no QA pipeline) |
/seer |
Sentry error investigation |
/summarize-channel |
Slack channel digest |
/clean_gone |
Clean deleted remote branches |
Rule: Plugin for quick action. Machine for pipeline with convergence.
Reference Skills (On-Demand Expertise)
Load domain knowledge into context when needed:
| Skill | Domain |
|---|---|
/ag-referencia-nextjs |
Next.js + React patterns |
/ag-referencia-typescript |
TypeScript strict mode |
/ag-referencia-python |
Python patterns |
/ag-referencia-supabase |
Supabase, PostgreSQL, RLS |
/ag-referencia-qualidade |
Quality gates, anti-theatrical tests |
/ag-referencia-sdd |
Spec Driven Development methodology |
/ag-referencia-seguranca-rules |
Security, LGPD, permissions |
/ag-referencia-mock-first |
Mock-First methodology for integrations |
Safety (Defense-in-Depth)
| Layer | What | How |
|---|---|---|
| PreToolUse | Block dangerous commands | force push, --no-verify, commits on main, config overwrites |
| PostToolUse | Detect secrets | Scans edited files for API keys, passwords, tokens |
| Rules (27) | Governance by context | Branch strategy, debug protocol, bulk change safety, memory limits |
| CLAUDE.md | Global behavior | Conventions, security, methodology |
Configuration
Customize CLAUDE.md
Copy CLAUDE.md to your project root and customize:
- Replace
[YOUR_PROJECT]with your project details - Add your stack-specific commands
- List your project's gotchas
Permissions (settings.local.json)
- Allow: npm, npx, bun, git, gh, node, and dev tools
- Deny: .env files, SSH keys, secrets, .pem/.key files
Adjust Model Routing
Edit agent frontmatter to change model assignments:
model: haiku # Fast, cheap — for scans and triage
model: sonnet # Balanced — for building and testing (default)
model: opus # Powerful — for complex analysis and specs
FAQ
Works with Python/Go/Rust/Java?
Yes. Agents are language-agnostic. Adjust typecheck/lint/test commands in your CLAUDE.md.
Do I need all 52 agents?
No. The 13 machines are the interface. Internal agents are called automatically. You never need to know they exist.
Works on Windows?
Yes. Shell hooks need Git Bash or WSL. Everything else works natively.
Can I customize agents?
Yes. Agents are plain .md files — edit freely. Use ag-melhorar-agentes to improve them based on error patterns.
What is QAT?
Quality Acceptance Testing. Instead of just checking "does it work?", QAT uses AI to evaluate "is the output good?" — scoring responses on domain-specific criteria.
Conflicts with other Claude Code setups?
No. Uses standard .claude/ directory structure. Compatible with other tools and plugins.
Stats
| Metric | Value |
|---|---|
| Sessions analyzed | 193+ |
| Commits generated | 464+ |
| Machines | 13 (ag-0 to ag-12) |
| Internal agents | 52 |
| Skills | 66 |
| Rules | 27 |
| Hooks | 8 (safety-only) |
| Playbooks | 12 |
| QAT scenarios | 173 |
| Plugins integrated | 24 |
| MCP servers | 8 |
| Reference skills | 8 |
Contributing
See CONTRIBUTING.md for guidelines on adding agents, skills, rules, and hooks.
Brand Assets & Design System
Logo vetorial + Design System oficial em assets/:
| File | What |
|---|---|
assets/logos/raiz-educacao-logo.svg |
SVG logo (pure paths, no fonts, any scale) |
assets/logos/raiz-logo-transparente.png |
PNG 900px, transparent background |
assets/logos/RaizLogo.tsx |
React/Next.js component |
assets/raiz-educacao-design-system.md |
UX/UI Design System (1045 lines, 13 sections) |
License
Built by Andre Gusman. Battle-tested across 193+ development sessions, 464 commits in 15 days.
Guide: Como Eu Trabalho com Claude Code Todo Dia
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found