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

Annotations already declare read-only and non-destructive behavior, so the description does not need to repeat that. It adds useful behavioral context by mentioning that results include graph_paths, DB original text, effective dates, and version status, and that long texts can be continued via get_evidence. However, it does not disclose any potential side effects, rate limits, or system-specific behaviors that go beyond the annotation.

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 verbose and not well front-loaded. The opening phrase '로컬 tax-ai-agent와 같은 통합 그래프 탐색' is vague and references an unspecified local agent. It then lists multiple output details and ends with pagination instructions. While it conveys necessary information, it could be condensed to focus on the tool's primary function and key differentiators, making it less concise than ideal.

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 the complexity of the search_intents object and the variety of document types, the description provides some high-level context (e.g., it returns graph paths and version info), but it lacks concrete examples of how to construct effective queries or interpret the response structure. The schema offers some guidance, but the description does not fully bridge the gap for an agent to autonomously use this tool in a nuanced scenario.

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?

The schema already provides descriptions for as_of, result_set_id, and search_intents, but the tool description does not add any meaningful explanation for limit, offset, or how to structure search_intents beyond what the schema offers. The mention of 'next_offset' refers to an output field, not a parameter, and 'get_evidence' is a separate tool. Thus, the description does not enrich parameter understanding beyond 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?

The description clearly states the tool's core function: an integrated graph search that traverses legal documents (law articles, precedents, rulings, interpretations, annexes, etc.) to collect issue-related materials. It distinctively positions itself as a comprehensive search tool compared to sibling tools that focus on single document types (e.g., get_article, search_cases). The phrase '쟁점 자료 수집의 우선 도구' (primary tool for collecting issue materials) further clarifies its intended role.

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?

The description provides some guidance by labeling this as the primary tool for collecting issue materials, implying it should be used first for comprehensive searches. However, it does not explicitly contrast it with sibling tools (e.g., when to use search_articles instead) or specify conditions for when this tool is not appropriate. The mention of supplementing with full-text search hints at a workflow but lacks clear directives.

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