convez.ai
MCP ServerProdutividadeAvançado
Ar9av

Agentic Tools Almanac

Ar9av/agent-manual

Guia de referência universal para configurar e estender agentes de IA

Ver no GitHub
InicialPluginScore · de 0 a 100

Manual completo e comparativo de hooks, configurações, MCP e skills para 17+ agentes de IA (Claude Code, Cursor, Copilot, Codex, Gemini CLI e outros). Referência técnica estruturada para interceptar eventos, customizar comportamentos e integrar Model Context Protocol em ferramentas agentic.

Instalação

$npx convez-ai Ar9av/agent-manual

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
15
Forks
1
Acompanhando
0
Saúde
28%
Último commit
há 2 meses
Versão
—
Issues abertas
4
Contribuidores
1

Casos de uso

  • Comparar hooks (PreToolUse, PostToolUse) entre diferentes agentes
  • Configurar MCP servers em Claude Code, Cursor, Copilot e demais ferramentas
  • Padronizar arquivos de instruções customizadas (.cursor/rules/, CLAUDE.md, CONVENTIONS.md)
  • Entender regras de exit-code, comportamentos de execução e descoberta de tools

Veredito

Recurso essencial e muito bem estruturado para desenvolvedores que trabalham com múltiplos agentes agentic. Preenche um vazio documentacional real ao centralizar configurações, hooks e padrões de 17+ ferramentas em um só lugar—recomendado para quem busca padronizar e otimizar fluxos multi-agent.

README

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

Agentic Tools Almanac ☤

The ultimate developer's index for AI coding agents. A community-driven reference for configuring, extending, and intercepting every major agentic development tool. Easily compare lifecycle hooks (e.g., PreToolUse and PostToolUse), built-in capabilities, Model Context Protocol (MCP) servers, system instruction file paths, and customized skills across 17+ different developer agents.


Tools Covered

ToolVendorConfig formatHooksMCPOfficial docs
Claude CodeAnthropicJSON✅ Full (31 events)✅code.claude.com/docs/en/hooks
Codex CLIOpenAITOML✅ Full (10 events)✅developers.openai.com/codex
GitHub CopilotGitHub/MicrosoftJSON✅ Full (13 events)✅docs.github.com/en/copilot/reference/hooks-reference
Gemini CLIGoogleJSON✅ Full (11 events)✅geminicli.com/docs
Factory DroidFactory.aiJSON✅ Full (9 events)✅docs.factory.ai
Kiro IDE/CLIAmazon (AWS)YAML✅ Full✅kiro.dev/docs
Kimi CodeMoonshot AITOML✅ Full — Beta (16 events)✅kimi.com/code/docs
CursorAnysphereJSON✅ Full (23 events)✅cursor.com/docs/hooks
HermesNousResearchYAML✅ Full (❓ events)✅hermes-agent.nousresearch.com
Devin CLICognition AIJSON✅ Full (7 events)✅cli.devin.ai/docs/extensibility/hooks
Pi Coding Agentearendil-worksJSON + YAML✅ Full (7 events)✅pi.dev
OpenCodeOpenCode.aiJSON✅ Via plugins✅opencode.ai/docs
OpenClawOpenClawYAML✅ Full (18 events via Plugin SDK)✅docs.openclaw.ai
TraeByteDanceMarkdown rules⚠️ Via MCP only✅docs.trae.ai
Trae CNByteDanceMarkdown rules⚠️ Via MCP only✅docs.trae.ai
AiderAider-AIYAML❌ Not shipped❌aider.chat/docs
Google AntigravityGoogleJSON / TOML✅ Full (5 events)✅antigravity.google

Legend: ✅ Full = shipped, documented hook system with blocking · ⚠️ = partial or workaround · ❌ = not available · ❓ = unknown


Quick Navigation

QuestionGo to
Which agents support activity X?_shared/activity-agent-matrix.md
What are all of Agent Y's hooks/tools/config?_shared/agent-tools-hooks-config.md
What is PreToolUse called in each tool?_shared/hook-event-comparison.md
Where does Tool Y store its config and skills?_shared/config-file-locations.md

Repository Structure

.
├── README.md                          # Master index + comparative benchmarks
├── tools/
│   └── <tool-name>/
│       └── README.md                 # Per-tool configuration & command references
├── _shared/
│   ├── activity-agent-matrix.md      # Capabilities matrix (natively supported tools)
│   ├── agent-tools-hooks-config.md   # Unified agent specifications & schemas
│   ├── hook-event-comparison.md      # Cross-agent hook event name lookup
│   ├── mcp-support.md                # MCP client capabilities & transport settings
│   └── config-file-locations.md      # Global/local paths & git-ignored overrides
└── _templates/
    └── tool-template.md              # Blueprint for documenting new developer tools

Hook Event Cross-Reference

How the same lifecycle moment is named across tools:

Lifecycle momentClaude CodeCodexGemini CLIGitHub CopilotKimi CodeFactory DroidCursorDevin CLIPi Agent
Before any toolPreToolUsePreToolUseBeforeToolpreToolUsePreToolUsePreToolUsepreToolUsePreToolUsetool.before.*
After any toolPostToolUsePostToolUseAfterToolpostToolUsePostToolUsePostToolUsepostToolUsePostToolUsetool.after.*
After tool failsPostToolUseFailure——postToolUseFailurePostToolUseFailure—postToolUseFailure——
Prompt submittedUserPromptSubmitUserPromptSubmitBeforeAgentuserPromptSubmittedUserPromptSubmitUserPromptSubmitbeforeSubmitPromptUserPromptSubmit—
Agent turn endsStopStopAfterAgentagentStopStopStopstopStop—
Subagent startsSubagentStartSubagentStart—subagentStartSubagentStart—subagentStart——
Session startsSessionStartSessionStartSessionStartsessionStartSessionStartSessionStartsessionStartSessionStartsession.created
Session endsSessionEnd—SessionEndsessionEnd—SessionEndsessionEnd——
Context compactedPreCompactPreCompactPreCompresspreCompactPreCompactPreCompactpreCompact——
Permission requestPermissionRequestPermissionRequest—permissionRequest———PermissionRequest—
LLM call——BeforeModel / AfterModel——————
NotificationNotification—Notificationnotification—Notification———

Blocking Mechanism Comparison

How each tool's PreToolUse-equivalent blocks a tool call:

ToolBlock methodStderr / reason goes to
Claude CodeExit 2LLM as context
Codex CLIExit 2 or {"decision":"block"} stdoutLLM
GitHub Copilot{"permissionDecision":"deny"} stdoutLLM (via permissionDecisionReason)
Gemini CLIExit 2LLM
Factory Droid{"permissionDecision":"deny"} stdoutLLM
Kimi CodeExit 2 or {"permissionDecision":"deny"} stdoutLLM
CursorExit 2 + failClosed optionLLM
Hermes{"action":"block","message":"..."} stdoutLLM
Devin CLIExit 2 or {"decision":"block"} stdoutLLM
KiroExit 2LLM
Pi AgentExit 2LLM

Source Confidence Key

Each tool README labels sources:

LabelMeaning
[official]Published by the vendor on their own docs site
[github]Vendor GitHub repo (README, source code)
[community]Community-maintained project (e.g., oh-my-pi) — may diverge from vendor
[installer-src]Inferred from the installer scripts or directory structures

Contributing

See _templates/tool-template.md for the standard page format.

When adding or updating a tool:

  1. Source everything to an official URL — no guessing.
  2. Mark uncertain items with ❓.
  3. Use the [official] / [github] / [community] labels in the Sources table.
  4. Update the master table in this README.
  5. Update _shared/agent-tools-hooks-config.md with the agent's section.

Data Sources & Disclaimer

The Hook Event Cross-Reference table and event counts in the Tools Covered table are sourced directly from each tool's official documentation at the time of last update. Hook APIs evolve rapidly — vendor docs are the authoritative source and should always be consulted before relying on counts or event names listed here. Entries marked ❓ could not be verified against live official documentation at the time of writing.

Ver o original no GitHub

Bom para

Relacionados