get_taste_profile

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

Annotations already establish readOnlyHint, idempotentHint, and destructiveHint=false, so the safety profile is covered. The description adds meaningful behavioral nuance beyond annotations: it states that bindings are per-project and do not transfer, it discloses that the returned data includes a full catalog and corpus, and it explicitly warns against treating it as a calibration step. This adds real context without contradicting 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 compact: one sentence states the operation and return value; a second sentence delivers the critical caveat and routing instruction. There is no redundant phrasing, no filler, and the most important distinguishing behavior ('NOT a calibration step') is prominent.

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?

Despite not having an output schema, the description enumerates the three components returned: rule catalog, precedent corpus, and per-project bindings. It also provides the key behavioral caveat about binding transfer and names the sibling to use instead when the binding is absent. For a single-parameter read-only tool, this is sufficient for an agent to call it correctly in 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?

Schema coverage is 100% and the only parameter 'name' is described simply as 'Profile name.' The description reinforces that loading is by name but adds no additional semantic detail about format, validation, or expected values. Baseline 3 is appropriate because the schema already carries the semantic load.

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 names a specific verb ('Load'), a specific resource ('locally stored taste profile by name'), and specifies the result contents (rule catalog, precedent corpus, per-project surface bindings). It also distinguishes itself from the calibration flow by explicitly stating what it is NOT, which sets it apart from get_taste_interview and list_taste_profiles.

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 provides an explicit alternative in get_taste_interview with a clear condition: when working on a project without a binding, call the interview instead of using this tool. It also warns that bindings are per-surface and do not transfer, which prevents misapplication. The context for when this tool is appropriate is clear from the contrast.

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