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

With no annotations, the description carries the full burden and does disclose a non-obvious behavior: chunks already sent in this session are not re-sent, which materially affects how an agent interprets results. It also reveals that result titles act as stable identifiers for re-fetching. It does not mention rate limits, result ordering, or relevance scoring, leaving some operational gaps.

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 scope and the natural-language/name input modes are front-loaded, and the two behavioral notes about title re-fetch and deduplication land at the end where they belong. The parenthetical list of execution-semantics examples is dense but each item illustrates a distinct query class rather than padding.

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 two-parameter lookup tool with an output schema, the definition covers purpose, input forms, and the session-deduplication behavior an agent would otherwise misread as missing results. The only meaningful omission is the meaning of 'limit', which is minor relative to the rest of the coverage.

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 0%, so the description must compensate, and it does so well for 'query' by explaining both accepted input forms and showing examples. The 'limit' parameter is never mentioned, so an agent gets no guidance on what it bounds or what the default of 5 implies, which is the one remaining semantic hole.

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 ('Search') and resource ('the manifoldbt documentation') and then enumerates the content covered (DSL rules, indicator/API signatures, config fields, execution semantics, metric interpretation, worked examples). That scope is clearly distinguishable from siblings like list_topics or explain_error, which handle navigation and error diagnosis rather than content lookup.

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 concrete use conditions with example queries in both natural language ('what happens when a bar gaps through my stop') and by name ('symbol_ref', 'warmup_bars'), and explains the follow-up pattern of passing an exact result title to re-read a chunk. It stops short of naming the sibling tools or stating when to prefer them over search_docs.

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