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

With no annotations, the description carries the full behavioral burden: it discloses the two execution modes and what each returns (value, probabilities, confidence, lifecycle ending vs. facts, composed verdict, receipt, anchored record) and warns that keyless mode ignores custom questions. It does not cover credential expectations for api_key, error behavior, or cost/limits, leaving meaningful gaps.

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 two-mode contrast is front-loaded, which is good structure, but the sentences are dense with unexplained jargon and the trailing 'A REF answer never produces a cheat' adds confusion rather than usable guidance. Some phrasing earns its place; some does not.

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?

No output schema and no annotations exist, so the description must explain both behavior and returns; it partially does so by enumerating the outputs of each mode. However, with 5 parameters at 20% coverage, undocumented params (api_key, options) and the mode/key relationship leave an agent unable to invoke it confidently.

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 only 20%, and 4 of 5 parameters are undocumented in the schema. The description touches typesafe_key (gates mode) and question (only asked in keyed mode), but api_key and options are never mentioned, and the relationship between api_key and typesafe_key is left ambiguous, so it does not adequately compensate for the coverage gap.

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 states a specific action (ask a question about an item / grade an item) and explicitly branches on whether a typesafe_key is present, naming the taifoon_judge_compose pipeline it maps to. This lets an agent tell the keyed and keyless modes apart, though the jargon ('calibrated answer', 'lifecycle ending', 'cheat') is opaque and the core resource noun ('item') stays abstract.

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?

It gives an explicit conditional: with your own typesafe_key you ask a custom question, without one you get the Taifoon grade and 'a custom question is not asked.' That routes the primary decision cleanly. It stops short of comparing against the many sibling judge/agent tools, so no explicit alternatives are named beyond the compose reference.

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