Skip to main content
Glama

agy_run

Send one authorized request to Gemini, Claude, or Codex through a supported channel, optionally overriding the model; missing CLI/login fails without retry, and runs may incur cost.

Instructions

Send one authorized request. participant: gemini/claude/codex (never client gpt); channel: antigravity/claude_code/codex_cli as supported. Default Gemini via Antigravity stays compatible. A channel override needs its explicit model; Claude requires Opus. Missing CLI/login fails without fallback/retry. May incur cost.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "prompt": "<prompt>"
    },
    "name": "agy_run"
  }
}

Input Schema

NameRequiredDescriptionDefault
modelNo
promptYes
channelNo
participantNogemini

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.8.0
    • addedInput schema / properties / channel
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Channel"
      +}
    • addedInput schema / properties / model
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Model"
      +}
    • addedInput schema / properties / participant
      Added value: +{
      +  "default": "gemini",
      +  "title": "Participant",
      +  "type": "string"
      +}
  2. First observedv0.5.2

TDQS

A3.5/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the annotations (openWorldHint=true, idempotentHint=false), the description discloses cost exposure, that missing CLI or login fails outright with no fallback/retry, and that a channel override imposes a model constraint. Those are real operational facts an agent cannot get from the schema; only the return/response behavior is left unstated.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Very short and front-loaded, with the default path stated before the override rules. The telegraphic phrasing ('participant: gemini/claude/codex') is dense but each clause carries distinct information, with no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 4-param, open-world, potentially cost-incurring tool with no output schema, the definition covers auth, cost, and model/channel rules but is silent on what the call returns and how it relates to the many agy_ discussion/council siblings. An agent can call it correctly but cannot predict its output shape or confidently choose it over agy_discussion_turn.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description carries the burden, and it does: participant enumerates gemini/claude/codex and excludes 'client gpt', channel enumerates antigravity/claude_code/codex_cli, and model is constrained by the channel override and Opus rules. Only 'prompt' is left to obvious inference, and cross-parameter coupling is spelled out.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose3/5

Does the description clearly state what the tool does and how it differs from similar tools?

The verb+resource ('Send one authorized request') plus the participant/channel vocabulary tells an agent this dispatches a single prompt to an external agent CLI. However, it never states what result comes back or how it differs from agy_discussion_turn or agy_council_round, so the agent must infer the one-shot vs. multi-turn distinction from the tool name alone.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives partial routing guidance — default Gemini via Antigravity stays compatible, channel overrides need an explicit model, Claude requires Opus — but gives no when-to-use/when-not-to-use versus siblings like agy_discussion_turn, and no statement that this is for single isolated calls rather than a multi-turn workflow.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.