GameCode MCP2

The GitHub copy is archived and no longer maintained.

A clean (and possibly naive) implementation of the Model Context Protocol (MCP) for tool integration.

Motivation - as few dependencies as possible, as simple and auditable a configuration as possible.

MCP is early technology. Allowing LLMs to execute system commands is inherently risky. This implementation prioritizes auditability over features - you can read every line that processes LLM requests. Even so, proceed with caution. Only time will tell if MCP's approach is sound.

- Direct tool exposure- Tools defined intools.yamlare exposed directly via MCP, not through meta-tools
- Clean protocol implementation- Pure JSON-RPC 2.0 over stdio without external dependencies
- Dynamic tool loading- Configure tools via YAML without recompiling
- Built-in and external tools- Support for both internal handlers and external commands

- mcp-server- The MCP server that loads tools fromtools.yamland exposes them via the protocol
- mcp-client- A client library for testing and integration

# Build the server cargo build --release --bin gamecode-mcp2 # Create a tools.yaml file (see examples/tools.yaml) cp examples/tools.yaml . # Run the server (it communicates via stdio) ./target/release/gamecode-mcp2

Add to your Claude Desktop configuration:

{ "mcpServers": { "gamecode": { "command": "/path/to/gamecode-mcp2" } } }

Themcp-clientcrate can be used as a dependency in gamecode-cli for MCP integration.

tools: - name: my_tool description: Description for the LLM command: /path/to/command # or "internal" for built-in args: - name: param1 description: Parameter description required: true type: string cli_flag: --param # null for positional internal_handler: handler_name # for internal tools

This implementation follows the MCP specification:

- initialize- Handshake with client
- tools/list- Returns all available tools
- tools/call- Execute a specific tool

Tools are exposed directly, not through meta-tools like "run".

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

The MCP server for Bitrix24 provides AI assistants with structured access to the Bitrix24 API. It delivers up-to-date method descriptions, parameters, and valid values, allowing assistants to work with precise data instead of guesswork. This reduces code errors and accelerates Bitrix24 integration development.

One remote MCP server for 500+ production APIs — Stripe, HubSpot, Postgres, Gmail, and more. OAuth and API key auth, credential management, and a CLI.

Single tool to control all 100+ API integrations, and UI components

Agent-native developer Q&A API with MCP + A2A endpoints for citations, job pickup, and answer submission.

Self-hosted MCP gateway: convert REST/SOAP/GraphQL/SQL APIs into MCP tools with 29 pre-built adapters, OAuth2, RBAC and audit log.

A universal bridge to convert any web API into an MCP server, supporting multiple transport types.

Dynamically creates MCP servers from web API configurations, integrating any REST API, GraphQL endpoint, or web service into MCP-compatible tools.

Hosted MCP server and coordination layer for AI coding agents — live API contracts, database schema, frontend/backend mismatch detection, and shared handoff tickets for Claude Code, Cursor, Codex, and Lovable.

An MCP server that dynamically loads tools from an external JSON file configured via an environment variable.