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

Annotations already declare readOnlyHint=false and destructiveHint=false; the description adds genuine value with the lifecycle fact that the judge "stays proposed until a person approves it" and that a human approval step is mandatory. It does not explain what happens on name collisions, whether an existing proposal is overwritten, or what the call returns for the proposed judge.

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?

Front-loaded with the verb, then use/do-not-use, next step, limits, and example in a compact sequence. The example is the weak link, since a config of {} illustrates required fields already enforced by the schema rather than the complex part.

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?

For a proposal-only tool with no output schema, the description covers lifecycle and the approval hand-off adequately. Given the complex nested schema and 20% description coverage, it is not complete enough for an agent to build a valid config without reading the raw schema closely.

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 description coverage is only 20% and the 5 parameters include a deeply nested config object (scope enum, judgeVersion hash, samplingRate, questions with pass_at/fail_at thresholds, scoreConfigs). The only parameter help is an example with an empty config ({}), which conveys nothing about the nested structure that carries nearly all the difficulty; "typed" hints at the question types but does not explain them.

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 and resource ("Propose a typed workflow or node judge") and scopes it to both workflow-level and node-level judges. It signals the proposal semantics that separate it from mutation siblings, but never names llm_judge_update, judge_propose, or workflow_judge_set, so an agent must infer the boundary from the verb alone.

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?

Gives an explicit trigger ("Use when an update needs quality checks"), an explicit exclusion ("Do not use when you need to approve or change a frozen judge"), and a stated follow-up (metrics_propose plus human approval). Strong when/when-not guidance, but it routes forward without naming the sibling tools that handle the excluded case.

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