capabilities

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

No annotations are supplied, so the description carries the full burden. 'Without loading a model' is a genuinely useful behavioral disclosure — it signals a lightweight, side-effect-free probe — but the description never explicitly states read-only semantics, latency, or whether any state is touched.

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?

One sentence, zero padding, with the key behavioral qualifier ('without loading a model') placed where it will be read. Nothing is wasted and nothing important is buried.

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 no explanation, and with no parameters the schema surface is trivial. The only remaining gap is that the description does not hint at what 'backends' or 'limits' concretely mean for a caller planning next steps.

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?

The tool takes zero parameters, which is the baseline-4 case. The description correctly describes it as a parameterless discovery call, and the empty schema corroborates this without contradiction.

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?

States a specific verb ('Discover') and a concrete resource set ('supported backends, operations and limits'), which no sibling tool covers — the rest of the catalog is run/study lifecycle management. It stops short of explicitly contrasting itself with any sibling, but the distinction is unambiguous in practice.

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 phrase 'without loading a model' implies the right context — cheap introspection before committing to a run — but the description never says when to call this versus proceeding directly to list_runs or submit_experiment, and offers no exclusions or prerequisites.

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