train_rlRun an allowlisted reinforcement learning recipe for arithmetic or math tasks to train models with group rollouts. Configures loss, group size, and PPO-style objectives while spending credits.
Run an allowlisted arithmetic/math group-rollout RL recipe. Spends credits.
| Name | Required | Description | Default |
|---|---|---|---|
| request | Yes | ||
| background | No |
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden of behavioral disclosure. It usefully notes that the operation 'Spends credits,' but says nothing about long-running behavior, background execution, permissions, side effects, or failure modes expected of a training job.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two front-loaded sentences with no wasted words: purpose first, cost warning second. It is efficiently structured, though its extreme brevity for such a complex tool borders on under-specification.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
An output schema exists, so return values need not be described. However, with no annotations, 0% schema coverage, nested objects, and many parameters, the description is far too thin to guide correct invocation; it omits usage conditions, parameter semantics, and most behavioral context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% and the tool has a deeply nested request object with many configurable fields. The description adds almost no parameter meaning beyond the words 'arithmetic/math' and 'group-rollout,' leaving the agent to infer the roles of model, recipe, training, evaluation, checkpointing, and background from property names alone.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb ('Run') and resource ('allowlisted arithmetic/math group-rollout RL recipe'), clearly identifying an RL training operation. It distinguishes itself from sibling training tools like train_sft and train_dpo by specifying RL, but it does not explicitly name alternatives or contrast with them.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given on when to use this tool versus alternatives such as training_start, recipe_start, train_sft, train_dpo, or train_distill. The phrase 'allowlisted arithmetic/math' hints at a constraint but does not tell the agent which scenarios call for this tool.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.