PluginBench
MCP Server
Active
MIT

io.github.PyModel/claude-agy-mcp MCP Server

io.github.PyModel/claude-agy-mcp

What is the io.github.PyModel/claude-agy-mcp MCP server?

Delegate heavy tasks from Claude Code to the Antigravity CLI (Gemini) with quota-aware failover.

How to install io.github.PyModel/claude-agy-mcp

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
Environment / auth
  • AGY_PATH

    Path to the agy binary (defaults to `agy` on PATH)

  • AGY_TIMEOUT

    Override the runtime ceiling for all tools, in seconds (use AGY_TIMEOUT_<TOOL> to override a single tool)

  • AGY_MAX_RUNTIME

    Absolute runtime ceiling in seconds (default 3600); the bridge does not kill on inactivity, only on cancellation, quota exhaustion, or this ceiling

  • AGY_MAX_OUTPUT_CHARS

    Truncation cap for tool output, in characters (default 50000)

  • AGY_DEFAULT_MODEL

    Fallback model when no chain entry is available (defaults to Gemini 3.6 Flash (High))

  • AGY_ON_FAILURE

    Set to `strict` to instruct the calling agent not to absorb failed delegations (default `fallback`)

~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "claude-agy-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@pymodel/claude-agy-mcp"
      ],
      "env": {
        "AGY_PATH": "<YOUR_AGY_PATH>",
        "AGY_TIMEOUT": "<YOUR_AGY_TIMEOUT>",
        "AGY_MAX_RUNTIME": "<YOUR_AGY_MAX_RUNTIME>",
        "AGY_MAX_OUTPUT_CHARS": "<YOUR_AGY_MAX_OUTPUT_CHARS>",
        "AGY_DEFAULT_MODEL": "<YOUR_AGY_DEFAULT_MODEL>",
        "AGY_ON_FAILURE": "<YOUR_AGY_ON_FAILURE>"
      }
    }
  }
}

Related MCP servers

UI design MCP: design references, intent routing, anti-slop static detector and gate. No API key.

View repository →

Private financial journal: receipts, recurring expenses, and cash flow with MCP tools.

View repository →
OUOuroboros logo

Ouroboros

Active

Agent OS that turns vague ideas into verified code through structured interviews, immutable specs, and replayable workflows.

5.6k
Python
MIT
View repository →
JMJMeter Docs logo

Apache JMeter community documentation, conversion tools, linters, and calculators for AI agents.

3
MDX
Apache-2.0
View repository →
DOdoiget logo

doiget

Active

OA-first paper fetcher: DOIs/arXiv IDs to local PDFs + metadata via Open Access APIs.

1
Rust
MIT
View repository →

MCP server for OpenMM — exposes market data, account, trading, and strategy tools to AI agents

View repository →