convez.ai
PluginDevOps & InfraestruturaIntermediário
agent-sh

Agnix

agent-sh/agnix

Valida configurações de agentes antes que quebrem seu workflow

Ver no GitHub
Muito bomPluginScore · de 0 a 100

Linter e validador para configurações de agentes AI, verificando SKILL.md, CLAUDE.md, hooks, MCP e outros arquivos de configuração. Previne que agentes quebrem silenciosamente devido a sintaxe incorreta, com 423 regras cobrindo Claude Code, Copilot, Cursor e mais. Oferece auto-fix e integração com VS Code, JetBrains, Neovim e Zed.

Instalação

$npx convez-ai agent-sh/agnix

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
490
Forks
35
Acompanhando
5
Saúde
87%
Último commit
há 1 dia
Versão
v0.61.0
Issues abertas
7
Contribuidores
14

Casos de uso

  • Validar SKILL.md e CLAUDE.md em CI/CD com GitHub Actions
  • Detectar misconfigurações em arquivos de hooks e MCP antes da execução
  • Garantir compatibilidade entre diferentes ferramentas de IA (Cursor, Claude Code, Copilot)

Veredito

Ferramenta essencial para quem trabalha com agentes AI multi-tool. Resolve um problema real (configs quebradas silenciosamente) com abrangência impressionante, mas requer familiaridade com arquivos de configuração de agentes.

README

A tradução para o português deste README ainda está na fila.

Why agnix?

Your skills don't trigger. Vercel's research found skills invoke at 0% without correct syntax. One wrong field and your skill is invisible.

"Almost right" is the worst outcome. 66% of developers cite it as their biggest AI frustration. Misconfigured agents produce exactly this.

Multi-tool stacks fail silently. Cursor + Claude Code + Copilot each want different formats. A config that works in one tool breaks in another.

Bad patterns get amplified. AI assistants don't ignore wrong configs - they learn from them.

agnix validates all of it - 423 rules sourced from official specs, academic research, and real-world breakage patterns. Auto-fix included.

Want to try it first? Open the playground - paste any agent config, see diagnostics instantly. No install, runs in your browser.

Quick Start

$ npx agnix .
Validating: .

CLAUDE.md:15:1 warning: Generic instruction 'Be helpful and accurate' [fixable]
  help: Remove generic instructions. Claude already knows this.

.claude/skills/review/SKILL.md:3:1 error: Invalid name 'Review-Code' [fixable]
  help: Use lowercase letters and hyphens only (e.g., 'code-review')

Found 1 error, 1 warning
  2 issues are automatically fixable

hint: Run with --fix, --fix-safe, or --fix-unsafe to apply fixes

https://github.com/user-attachments/assets/72d5fe7c-476f-46ea-be64-5785cf6d5600

Install

# npm (recommended, all platforms)
npm install -g agnix

# Homebrew (macOS/Linux)
brew tap agent-sh/agnix && brew install agnix

# Cargo
cargo install agnix-cli

Pre-built binaries | All install options

Editor Extensions

EditorInstall
VS CodeVS Code Marketplace
JetBrainsJetBrains Marketplace
Neovim{ "agent-sh/agnix", config = function() require("agnix").setup() end }
ZedSearch "agnix" in Extensions

Editor setup guide

GitHub Action

- name: Validate agent configs
  uses: agent-sh/agnix@v0
  with:
    target: 'claude-code'

Usage

agnix .              # Validate current directory
agnix --fix .        # Apply HIGH and MEDIUM confidence fixes
agnix --fix-safe .   # Apply only HIGH confidence fixes
agnix --fix-unsafe . # Apply all fixes, including LOW confidence
agnix --dry-run --show-fixes .  # Preview fixes with inline diff output
agnix --strict .     # Strict mode (warnings = errors)
agnix --target claude-code .  # Legacy target preset (primarily affects CC-* rules)
agnix --target kiro .         # Legacy target preset (use tools = [...] for tool-only filtering)

Full CLI reference | Full rules reference

Supported Tools

ToolRulesCountConfig Files
Agent SkillsAS-*, CC-SK-*31SKILL.md
Claude CodeCC-*53CLAUDE.md, hooks, agents, plugins
GitHub CopilotCOP-*6.github/copilot-instructions.md, .github/instructions/*.instructions.md
CursorCUR-*16.cursor/rules/*.mdc, .cursorrules, .cursor/hooks.json, .cursor/agents/**/*.md, .cursor/environment.json
KiroKIRO-*, KR-SK-*, KR-AG-*, KR-HK-*, KR-PW-*, KR-MCP-*52.kiro/steering/**/*.md, .kiro/skills/**/SKILL.md, .kiro/agents/*.json, .kiro/hooks/*.kiro.hook, .kiro/settings/mcp.json, .kiro/powers/*/POWER.md
MCPMCP-*12*.mcp.json
AGENTS.mdAGM-*, XP-*13AGENTS.md, AGENTS.local.md, AGENTS.override.md
ClineCLN-*4.clinerules, .clinerules/*.md, .clinerules/*.txt
Gemini CLIGM-*9GEMINI.md, GEMINI.local.md, .gemini/settings.json (hooks), gemini-extension.json (extensions), .geminiignore

Architecture

Rust workspace crates:

  • agnix-rules - rule metadata generated from knowledge-base/rules.json
  • agnix-core - shared validation engine library
  • agnix-cli - command-line interface binary
  • agnix-lsp - language server binary
  • agnix-mcp - MCP server binary
  • agnix-wasm - WebAssembly bindings for browser/runtime integrations

Contributing

Contributions welcome. See CONTRIBUTING.md for the development guide.

Report a bug | Request a rule | Good first issues

License

MIT OR Apache-2.0


Ver o original no GitHub

Bom para

Relacionados