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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, and the description adds substantive behavioral details beyond those: the embedding model (BGE-base-en), similarity function (cosine), windowing strategy (500-char overlapping windows), and hard input cap (200K chars, longer truncated and flagged). It also reveals that results include offsets and scores, which is actionable for an agent needing to verify quotes. No contradiction exists between description and 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 dense but every sentence earns its place. It opens with the core action, then explains the value proposition (saves context), technical specifics (embeddings, windowing, cap), and a pairing suggestion. There is no filler; the structure moves from what → when → how → caveats. Despite its length, it is well-organized and front-loaded with the most critical information.

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?

Given three parameters, no output schema, and a moderately complex behavior, the description covers all necessary context: input semantics, examples, output characteristics (passages, offsets, scores), performance boundaries (200K cap, truncation flag), and a use-case workflow with ask_pipeworx_grounded. There is no missing critical detail for an agent to decide whether to call and what to expect back. The absence of an output schema is mitigated by the clear description of the return content.

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 100%, so baseline is 3. The description adds notable extra semantics: it explains that 'text' means content already pulled (e.g., SEC 10-K body, an article), gives concrete query examples, and clarifies that the return includes character offsets so the agent can verify verbatim quotes. This goes beyond the basic field descriptions, though it doesn't deeply detail the exact JSON shape of the output. A 4 is justified for the added context and examples.

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 opens with a clear verb-resource pair: 'Semantic search INSIDE a fetched record.' It explicitly details the inputs (text + query) and outputs (passages with offsets and similarity scores), and it distinguishes itself from sibling tools by emphasizing that it operates on already-fetched content rather than fetching or grounding over whole documents. This is a specific, well-differentiated purpose statement.

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?

Provides explicit when-to-use guidance: 'Use when the record is too big to cram into the prompt' and directly names a complementary alternative: 'Pairs with ask_pipeworx_grounded: fetch with the gateway, ground over the relevant passages instead of the whole document.' This gives an agent clear decision criteria and a workflow context, far exceeding a simple 'use this for search.'

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