미공시 내부거래 교차탐지 (J004↔J001)

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

With zero annotations the description carries the full burden and does so: it discloses the summary-vs-detail split (up to 1MB via read_detection_result), the obligation to pass required_warnings/summary_incomplete/details_required through unchanged, a ~50-second self-stop yielding partial results, 20 companies per call, continuation semantics, and the not_judged category. It also warns about false-positive causes (한도성 의결, 약관 금융거래 특례) and the cost of misjudgment via penalty figures.

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?

Front-loaded with the core purpose and then organized into bulleted constraints with ★ markers for must-reads, so the density is largely earned given the tool's pitfalls. It is still long and partially restates the continuation_token schema text, so it is not maximally tight.

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, yet the description names the key response fields an agent must handle (action_items_preview, required_warnings, summary_incomplete, details_required, continuation.complete, continuation.stalled, not_judged, scope_caveats). For a complex cross-document detector this covers what an agent needs to call and interpret it 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 behavior beyond the schema: continuation_token must be paired with the same rcept_no/group, previously fetched J001 lists are not refetched, and each call covers 20 companies. It also reinforces the year-vs-prior-fiscal-year relationship, though that is largely duplicative of 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?

States a precise verb+resource: cross-matching actual transaction tables in the J004 annual representative-company form against J001 large-scale internal-transaction disclosures to surface 'undisclosed candidates.' It explicitly frames outputs as candidates, not confirmations, which distinguishes it from audit/consistency siblings like check_j004_consistency and audit_group_disclosures.

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 rich operational guidance: read action_items_preview first, an empty list does not mean 'no issue', follow continuation_token until complete:true and present only the last call's result, and read details via read_detection_result. It names the detail-retrieval sibling but never explicitly contrasts when to pick this over other audit tools, so it stops short of full when/when-not coverage.

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