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

With no annotations, the description carries the full burden and does so thoroughly: it discloses case-insensitive literal substring matching, the fields searched, empty-query behavior, ordering by partition then task ID, explicit pagination, no relevance scoring, and the nuance that the train subset is not double-counted. These are non-obvious behavioral traits that help the agent anticipate results.

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 tight and front-loaded: the first sentence states the core purpose, followed by clarifying behaviors. Each sentence adds unique value—empty-query behavior, allowed values, ordering/pagination, and the train double-count caveat. No redundancy or filler.

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?

Given a search tool with five parameters and no output schema, the description covers the essential operational details: what is searched, how to filter, ordering, pagination, and an edge case about split counts. It implicitly indicates the return type (tasks) and is sufficient for an agent to invoke correctly.

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 coverage is 0%, so the description must compensate. It explains the meaning of query (substring), partition ('all', 'shellops', 'shellops_pro'), and split ('all', 'train_src', 'train', 'test'), and implies limit/offset via 'explicit pagination'. It does not explicitly define limit/offset constraints, but given conventional naming, the added meaning is substantial beyond the empty schema descriptions.

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 begins with a specific verb ('Find') and a clear resource ('real ShellOps CLI benchmark tasks'), then defines the search scope (substring in complete instruction, task ID, or published task type). This distinguishes it from siblings like Agentic_RL_search_evidence (searches evidence) and Agentic_RL_get_task (retrieves a specific task).

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?

Provides clear context on how to filter results via partition and split, and explains ordering and pagination behavior. However, it does not explicitly state when to use this tool versus siblings (e.g., 'use get_task when you have a task ID'), though the verb and resource imply the use case. There are no exclusions or alternative routing.

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