force_training

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

With no annotations, the description carries the full behavioral disclosure burden. It reveals that the training runs as a background process, that it always overwrites previous results for the same run_id, and that it blocks until ready with mode-specific conditions. It stops short of explaining process lifecycle details such as whether launch failures clean up partial processes, but the disclosed behaviors are important and clearly stated.

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 well structured: purpose and critical warning first, then blocking behavior, then a clean Args list. No sentence is purely decorative, and the overwrite warning is front-loaded where an agent will see it. The Args list is somewhat long, but it is necessary given the zero-coverage schema.

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?

For an 11-parameter launch tool with no annotations, the description is fairly complete. It covers all parameters, explains the blocking semantics in both modes, and highlights the destructive overwrite behavior. Since an output schema exists, omitting return-value details is acceptable. A little more context about post-launch monitoring or failure cleanup would round it out, but the core operational picture is present.

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%, so the description must compensate for the schema's silence. It does this by providing a one-line semantic explanation for every parameter, including useful details like torch_device example values, base_port auto-assignment, seed -1 meaning random, and the wait behavior split by mode. It could add more nuance, such as constraints on time_scale or port ranges, but it meaningfully enriches all parameters.

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 opens with a specific verb and resource: 'Launch a new mlagents-learn training run as a background process.' It clearly identifies what the tool does and immediately distinguishes it from resume_training by warning about overwriting previous results for the same run_id. This gives an agent a precise mental model of the tool's role.

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?

The description provides explicit when-to-use and when-not-to-use guidance: 'To continue from a checkpoint, use resume_training instead.' It also explains the two modes (editor vs batch) and the blocking behavior in each, which helps an agent decide how to invoke it and what to expect. This is strong, actionable usage guidance.

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