Agent_Exchange_MCP

An MCP server that lets an AI client (Claude Code, Claude Desktop, …) talk to other AI models through their APIs. Supports Anthropic, OpenAI and xAI, with multi-turn conversations saved to disk.

Features

• Multiple providers — Anthropic (Claude), OpenAI, xAI (Grok). Choose the provider from the client UI, and switch it per call, even mid-conversation.

• Multi-turn conversations — history is kept per conversation_id.

• Persistent — each conversation is a JSON file on disk and survives server restarts.

• Simple config — API keys and models live in a .env file.

Related MCP server: Multi-CLI MCP

Installation

Requires Python 3.11+.

git clone https://github.com/cat-people-82/Agent_Exchange_MCP.git
cd Agent_Exchange_MCP
pip install -e .
cp .env.example .env    # then add your API keys

Configuration

Settings are read from .env in the project root (or the current directory, or the file named by AGENT_EXCHANGE_ENV_FILE). Real environment variables take precedence over .env. .env is git-ignored — never commit it.

Provider selection is not configured here — it happens in the client (see below). .env only holds API keys, models and the data directory. You only need keys for the providers you use. Adjust the default OpenAI/xAI model IDs to whatever your account offers.

Register with an MCP client

Claude Code:

claude mcp add agent-exchange -- agent-exchange-mcp

Claude Desktop — edit claude_desktop_config.json (Settings → Developer → Edit Config). command must be an executable (not the path to server.py), so point it at the Python interpreter of the environment where you ran pip install -e . and run the package as a module:

{
  "mcpServers": {
    "agent-exchange": {
      "command": "/absolute/path/to/Agent_Exchange_MCP/.venv/bin/python",
      "args": ["-m", "agent_exchange_mcp"]
    }
  }
}

Tips:

• Create the environment first: python3 -m venv .venv && .venv/bin/pip install -e . (find an existing interpreter's path with which python while its environment is active).

• Claude Desktop doesn't inherit your shell environment; keys are read from the project's .env, or pass them via an "env": { "ANTHROPIC_API_KEY": "..." } entry.

• Fully quit and reopen Claude Desktop after editing the config. Logs are in ~/Library/Logs/Claude/mcp-server-agent-exchange.log.

Tools

chat returns {conversation_id, provider, model, reply}.

Example flow: ask Claude a question, then continue the same conversation with provider: "xai" to get Grok's take with the full history.

Choosing a provider

The provider is picked in your MCP client, not in config:

• chat has a provider parameter (anthropic | openai | xai), which clients show as a choice list and which the calling model or you can set.

• If it is omitted when starting a conversation and more than one provider has an API key, the server asks you to choose through the client's prompt (MCP elicitation, supported by clients such as Claude Code).

• If the client doesn't support elicitation, or only one provider has a key, the first provider with a key is used.

• A continued conversation keeps its last provider unless you pass a different one.

Storage

Conversations are stored as <conversation_id>.json (system prompt + message list) in the data directory. Files are plain text and contain your prompts and replies, so protect the directory accordingly. Delete a conversation with reset_conversation or by removing its file.

Troubleshooting

• has_api_key: false — call server_info to see which .env files were checked. Sandboxed clients may not let the server read the project folder; set the keys in the client's environment-variable settings instead. Config is read at startup, so reconnect after changes.

• Could not reach the <provider> API: Connection error — the client app may export proxy variables (ALL_PROXY, HTTPS_PROXY) that point at a proxy this process can't reach. server_info shows them under proxy_env. If the machine can reach the API directly, set AGENT_EXCHANGE_IGNORE_PROXY=true.

• Using SOCKS proxy, but 'socksio' is not installed — reinstall (pip install -e .); socksio is a dependency.

• SSLCertVerificationError: ('OSStatus -26276',) (macOS) — the SDK default client verifies through the OS trust store (httpx 2.x uses truststore, which calls the Security framework) and that can fail even for a valid public certificate. Both client factories therefore verify against certifi; override with AGENT_EXCHANGE_CA_BUNDLE if you sit behind a TLS-terminating proxy whose CA is not in certifi.

• A sandboxed client may offer no direct egress. If AGENT_EXCHANGE_IGNORE_PROXY=true produces nodename nor servname provided, the proxy is the only route out — leave the flag off. Where ALL_PROXY is SOCKS but only the HTTP proxy serves, the server rewrites the catch-all to the HTTP proxy at startup and reports it as socks_fallback in server_info; disable with AGENT_EXCHANGE_SOCKS_FALLBACK=false.

Notes

• If a provider call fails, the turn is not saved, so history stays consistent.

• Anthropic calls use adaptive thinking and streaming; OpenAI and xAI use the Chat Completions API.