RCF MCP Server

Overview Schema Source Related Servers Discussions Admin

compute.plan_run

compute_plan_run

Read-only

Build a project execution plan with node selection, environment checks, GPU choice, and the exact command—without starting, writing, or reserving anything.

Instructions

Build an execution plan for a project (node selection, environment check, GPU choice, exact command). Pure planning: nothing is started, written or reserved. Optional pilot_seconds uses a hard watchdog when submitted; runtime_budget_seconds only gates planning/alerts, never cancels a normal run. Neither walltime nor budget is an ETA.

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": {
      "project_id": "<project_id>"
    },
    "name": "compute_plan_run"
  }
}

Input Schema

Name	Required	Description	Default
`cpus`	No
`gpus`	No
`tasks`	No
`env_id`	No
`params`	No
`node_id`	No
`memory_gb`	No
`partition`	No
`project_id`	Yes
`time_minutes`	No
`gpu_memory_gb`	No
`pilot_seconds`	No
`allow_shared_gpu`	No
`command_override`	No
`runtime_budget_seconds`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Nov 7, 2026

TDQS

A4.1/5.0

Behavior5/5

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

Annotations only declare readOnlyHint=true, yet the description adds genuinely new behavioral detail: pilot_seconds installs a hard watchdog when submitted, runtime_budget_seconds gates planning and alerts but never cancels a normal run, and neither walltime nor budget is an ETA. These are non-obvious semantics an agent could not derive from the schema or annotations.

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?

The purpose leads, the no-side-effects guarantee follows, and the two timing caveats are packed into one dense sentence with zero filler. The middle sentence is compressed to the point of being jargon-heavy, but nothing is redundant.

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 15-parameter planning tool with no output schema, the description omits what the returned plan actually looks like and how the many resource parameters interact or default. The safety and timing model is well covered, but an agent still has to guess at most of the request surface.

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

Parameters2/5

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

Schema description coverage is 0% across 15 parameters, so the description carries the full explanatory burden. It clarifies only two of them (pilot_seconds, runtime_budget_seconds); the other thirteen — cpus, gpus, tasks, params, node_id, memory_gb, partition, time_minutes, gpu_memory_gb, allow_shared_gpu, command_override, env_id — remain undocumented anywhere, including whether they are request hints versus hard constraints.

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

Purpose5/5

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

"Build an execution plan for a project" gives a specific verb and resource, and the parenthetical enumerates the plan's contents (node selection, environment check, GPU choice, exact command). Saying it is "pure planning" separates it cleanly from compute_submit_run and the apply_* siblings without needing their schemas.

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

Usage Guidelines4/5

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

"Pure planning: nothing is started, written or reserved" is an explicit when-to-use rule that routes the agent to compute_submit_run for real execution and to this tool for a dry-run preview. It does not name that alternative outright, so the choice must be inferred from context rather than stated.

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