开源MCP工具

Requirements

You need at least one provider:

• Codex CLI (GPT): npm install -g @openai/codex, then codex login.
• Antigravity CLI: Getting Started with Antigravity CLI and Migrating from Gemini CLI, then run agy and login.
• Grok (xAI): no CLI to install; the bridge ships with the plugin (needs Node 18+). Set XAI_API_KEY (get a key at https://console.x.ai).
• OpenRouter: no CLI; the bridge ships with the plugin (needs Node 18+). Set OPENROUTER_API_KEY (get a key at https://openrouter.ai/keys), then declare models in ~/.config/deliberation/config.json (Windows: %APPDATA%\deliberation\config.json; override with DELIBERATION_CONFIG). Works with any OpenAI-compatible endpoint (Ollama, vLLM, LM Studio, HuggingFace Inference) - auth is skipped automatically when the key env var is empty.

Install

How to Use

Describe your task. Claude detects when an expert helps and delegates automatically:

You: "Is this authentication flow secure?"
Claude: routes to the Security Analyst, then synthesizes the findings.

You can also ask explicitly: "Ask GPT to review this architecture", "Ask Gemini to...", or "Ask Grok to...". Each expert runs read-only for analysis or with write access to apply fixes, and Claude picks the mode from your request.

Or invoke the slash commands directly - see Commands above.

Configuration

Every expert supports two modes, chosen automatically from your request:

OpenRouter config

OpenRouter models are declared in ~/.config/deliberation/config.json - the canonical XDG path (Windows: %APPDATA%\deliberation\config.json). You can override the path with DELIBERATION_CONFIG. The file is the live single source of truth: changes to models, routing, or the providers.openrouter block hot-reload without restarting Claude Code. Toggling a built-in provider (codex / gemini / grok) still requires /setup.

The config has four sections: providers (transport / connection per provider), models (named model records keyed by id), routing (fan-out policy), and consensus.arbiter (who synthesizes the verdict). The $schema key gives editors validation and autocomplete - VS Code needs no extension.

Minimal example:

{
  "$schema": "https://raw.githubusercontent.com/antonbabenko/deliberation/master/config/config.schema.json",
  "version": 1,
  "providers": {
    "codex":  { "enabled": true },
    "gemini": { "enabled": true },
    "grok":   { "enabled": true, "apiKeyEnv": "XAI_API_KEY" },
    "openrouter": {
      "enabled": true,
      "apiKeyEnv": "OPENROUTER_API_KEY",
      "apiBase": "https://openrouter.ai/api/v1",
      "defaultModel": "openai/gpt-4.1-mini",
      "defaults": { "reasoningEffort": "medium" }
    }
  },
  "models": {
    "gpt-4-or": {
      "provider": "openrouter",
      "model": "openai/gpt-4.1",
      "askAll": true,
      "consensus": false
    },
    "claude-arb": {
      "provider": "openrouter",
      "model": "anthropic/claude-haiku-4-5",
      "askAll": true,
      "consensus": true,
      "reasoningEffort": "high"
    }
  },
  "routing": { "maxFanout": 3 },
  "consensus": { "arbiter": { "model": "claude-arb" } }
}

Browse model slugs at openrouter.ai/models; the model field takes any slug listed there. Each record's provider must be "openrouter" in v1 (codex / gemini / grok are managed by their own CLI / API).

reasoningEffort (low / medium / high) sets how hard a reasoning model thinks. Put it on providers.openrouter.defaults to cover every model, or on a single record to override the default for that one. Precedence runs call argument over per-record override over defaults.

/ask-all includes records where askAll !== false, capped to routing.maxFanout. /consensus includes records where consensus === true, with no fanout cap (a warning is emitted when more than 3 models participate). consensus.arbiter picks who synthesizes: a shorthand string ("auto" / "host" / "codex" / "gemini" / "grok") or { "model": "<id>" } naming a record (even an out-of-panel one). Implementation tasks always route to Codex or Gemini - never OpenRouter.

For the full schema, the $schema / VS Code validation story, apiBase override matrix (Ollama, vLLM, LM Studio, HuggingFace), file-attachment caps, session model persistence, consensus cost model, and error kinds, see TECHNICAL.md - OpenRouter bridge.

For provider defaults, environment variables, and manual MCP setup, see TECHNICAL.md.

Claude Code plugin (recommended):

1. Add the marketplace - antonbabenko/agent-plugins

/plugin marketplace add antonbabenko/agent-plugins

2. Install the plugin

/plugin install deliberation@antonbabenko

3. Run setup

/deliberation:setup

Claude now routes complex tasks to your GPT, Gemini, Grok, and OpenRouter experts (Grok and OpenRouter advise; GPT and Gemini can also implement).

Alternative: Use `deliberation` MCP server (standalone, works with any agents)

The orchestration server is also published on its own - npm @antonbabenko/deliberation-mcp, Official MCP Registry name io.github.antonbabenko/deliberation.

One-click install:

<details> <summary>Manual config for any MCP clients</summary>

Add this to your host's MCP config (most hosts use the mcpServers key):

{
  "mcpServers": {
    "deliberation": {
      "command": "npx",
      "args": ["-y", "@antonbabenko/deliberation-mcp"],
      "env": {
        "XAI_API_KEY": "xai-...",
        "OPENROUTER_API_KEY": "sk-or-v1-..."
      }
    }
  }
}

The env block is how you set provider keys outside Claude Code. GPT and Gemini do not read keys here - they use the codex and agy CLIs (logged in separately), so drop those lines if you only use GPT/Gemini. XAI_API_KEY enables Grok; OPENROUTER_API_KEY enables OpenRouter (which also needs models declared in ~/.config/deliberation/config.json - the canonical XDG path, Windows %APPDATA%\deliberation\config.json - or point elsewhere with DELIBERATION_CONFIG). The one-click buttons above cannot carry secrets - add the env block by hand after installing.

Per-host config location and the key it expects:

Provider prerequisites are the same as the plugin (see Requirements): the Codex CLI for GPT, agy for Gemini, XAI_API_KEY for Grok, and OPENROUTER_API_KEY plus ~/.config/deliberation/config.json for OpenRouter (Windows: %APPDATA%\deliberation\config.json; override the config path with DELIBERATION_CONFIG).

Tools exposed: ask-all, consensus, ask-gpt / ask-gemini / ask-grok / ask-openrouter, and the seven experts (architect, plan-reviewer, scope-analyst, code-reviewer, security-analyst, researcher, debugger).

The package also ships a deliberation-setup bin. Run it once with npx -y --package @antonbabenko/deliberation-mcp deliberation-setup to write a starter ~/.config/deliberation/config.json (it never overwrites an existing one). The plain npx -y @antonbabenko/deliberation-mcp form runs the default bin (the server), which is what your MCP host launches. For host rule wiring, see AGENTS.md and the per-host snippets in examples/.

</details>