Prepare a non-executing MCP selection plan

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint. The description adds behavioral context by stating it preserves unresolved dimensions, rejects credential fields, and never authorizes or executes, which goes beyond the annotations.

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 brief (two sentences) but the first sentence is dense with jargon, making it less clear. It lacks a clear breakdown of what the tool does in simpler terms.

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 the complexity (5 parameters, no output schema, nested objects), the description lacks details on the return value or how the plan is used afterward. It does not cover what the output contains, leaving the agent with insufficient 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?

The description refers to 'audience, purpose and caller-declared expiry,' linking to three parameters, but does not explain 'arguments,' 'record_id,' or their semantics. With 0% schema coverage, the description should provide more parameter detail.

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 it binds a frozen identity and snapshot digest to tools, audience, purpose, and expiry, and emphasizes it never executes. The title 'Prepare a non-executing MCP selection plan' reinforces this. However, jargon like 'frozen identity' and 'snapshot digest' may not be universally understood, slightly reducing clarity.

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 implies it should be used for preparing a plan without execution ('Never authorises or executes a query') and mentions preserving unresolved dimensions and rejecting credentials. It does not explicitly contrast with siblings like okf.get_record or okf_eval.submit_answer, but the distinct roles are inferable.

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