Embeddings

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

Annotations declare readOnlyHint=false, idempotentHint=false and openWorldHint=true, which are otherwise confusing for a 'compute' tool; the description supplies the missing reason — 'Paid per input token' — so the agent understands each call costs money and is not a free repeatable read. It also discloses that route 'carpe' is always served in a TEE while 'router' may be served outside one. It stops short of describing the response shape or batching behavior.

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?

Three short sentences, front-loaded with the core action before the routing and cost caveats. Dense but readable; 'privacy applies' is terse to the point of being cryptic, costing a point.

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?

Covered: what it does, batch limit, cost model, routing/privacy trade-off. Missing for a 6-parameter tool with no output schema: any indication of what comes back (vector arrays, dimensions of the returned vectors, base64 encoding implications) and the default model behavior. Adequate but leaves real gaps.

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 67%, so a baseline 3 applies. The description restates what the schema already documents for `route` (carpe vs router, cheapest market) and the 1–128 input count, and adds nothing about `model`, `dimensions`, or `encoding_format`; the privacy semantics it gestures at ('privacy applies') are only spelled out in the schema.

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 ('Compute embedding vectors') plus the compatibility contract (OpenAI-compatible) and the batch bound (1–128 texts). That is unambiguous and clearly separable from siblings like carpe_chat, carpe_tts, or carpe_list_models.

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?

Usage context is implied rather than stated: the agent can infer this is the text-to-vector tool and that 'router' is the cheapest option with privacy trade-offs. There is no explicit when-to-use/when-not guidance versus alternatives, and no pointer to carpe_list_models for discovering model ids (that hint lives only in the schema).

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