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

Annotations only cover the read-only/idempotent safety profile, and the description adds substantial context beyond that: per-call pricing ($0.001 USDC via x402 on Base), that errors are never charged, hard input limits (100 inputs, 8000 chars each, 60000 total), and that the GET form with no params embeds a sample sentence. This is exactly the operational detail an agent needs before invoking a paid endpoint.

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?

Pricing is front-loaded in brackets, followed by the request/response shape, then models, limits, and use cases in a logical order. It is dense and long, but every clause carries information; the model enumeration is the only part that could be tightened.

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?

With no output schema, the description compensates by spelling out the response object ({object: "list", data: [{embedding, index}], model, usage}). Combined with the documented limits, pricing, and model aliasing, an agent has everything needed to call and interpret results correctly.

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%, so the baseline is 3, but the description adds real meaning: `dimensions` 'truncates and re-normalizes' (more precise than the schema's wording) and any unrecognized model name silently maps to bge-m3 with the actual model reported in the response. The model-alias fallback is behavior no schema field conveys.

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+resource ('Drop-in OpenAI Embeddings endpoint') with the exact request/response contract, immediately distinguishing it from siblings like rerank_documents and llm_chat. An agent can tell what it produces (vectors) and in what format without opening the schema.

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?

Names concrete use cases ('For RAG, semantic search, clustering and dedupe') and clarifies the default behavior when no params are given, so the agent knows when this is the right tool. It stops short of explicitly routing between this and rerank_documents or llm_chat, so no when-not/alternative guidance.

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