sw_export_eval_template

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

With no annotations, the description fully discloses behavior: writes code to a .py file, replaces the my_algorithm function, runs the script, registers two arms, prints result_id, and clarifies that MCP does not execute user code and that per-sample values go to .npz rather than MCP JSON. This is highly transparent.

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 with numbered steps and bolded key notes. Despite length, every sentence adds value, providing necessary operational details without redundancy.

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?

The description covers the workflow, return value (result_id), and file output (.npz), which is good given the output schema exists. However, it omits the roles of dataset_id and metric, leaving a gap in understanding the tool's inputs.

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 schema has two parameters (dataset_id, metric) with 0% description coverage, and the description fails to explain either. Worse, it references a 'code' parameter that is not present in the schema, creating confusion about the actual input contract.

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 clearly states the tool exports an evaluation script skeleton for custom algorithms, explicitly contrasting with sw_compare_arms which only supports six precoding types. It specifies the verb (export), resource (eval script skeleton), and scope (custom algorithms), distinguishing it from siblings.

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 explicit when-to-use guidance: for custom algorithms not supported by sw_compare_arms. It gives a numbered step-by-step workflow, including how to replace the algorithm, run the script, and hand off result_ids to sw_compare_results, plus a note about MCP not executing user code.

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