search_budget_projects

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

Annotations only cover the safety profile (readOnly, non-destructive, closed-world), so the description adds substantial extra context: row count (326289), known coverage gaps, the reconciliation each row passed, and the crucial unit clarification that amounts are thousand-KRW appropriations rather than benefits or spending. It also sets a citation obligation ('Cite source plus PDF page or HWP table/row; never fabricate a page') and warns that an empty result is not proof of absence — exactly the behavioral caveats an agent needs.

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?

Purpose is front-loaded in the opening sentence and essentially every clause carries load-bearing information (coverage limits, units, citation rules). The telegraphic, comma-spliced style is dense to the point of being hard to parse, but nothing is redundant or filler.

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?

There is no output schema, so the description has to characterize results itself — and it does, describing what a row represents, the unit of measure, and the reconciliation provenance. Pagination behavior (limit max 20 / offset) is not described, which is the main remaining gap for a 4-parameter search tool.

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 0%, so the description carries the full burden, and it explains the two non-obvious parameters well: q is 'space-separated AND business-name keywords' and institution 'filters region' (a non-obvious mapping given the parameter name). Only limit/offset are left implicit, which is acceptable for standard pagination, so the meaningful gap is small.

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 first sentence gives a specific verb (Search) plus a tightly scoped resource: 'indexed 2026 original-budget general-account BUSINESS expenditure rows.' An agent can tell this apart from the totals/notice/document siblings by the 'indexed original-budget rows' scope. It stops short of naming the nearest sibling (search_verified_budget_totals) explicitly, so differentiation is inferential rather than stated.

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 names one alternative with a condition ('Use research_policy for plans/execution') and supplies caveats about coverage ('some municipalities and departments are not covered', 'No hit does not prove absence'). However, it never says when to prefer this over the closely-related search_verified_budget_totals, leaving the most important routing decision to inference.

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