Pixiv 插画搜索

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

With no annotations, the description carries the full burden and does so thoroughly: it discloses blacklist rejection of the entire request, automatic filters (mutual-exclusive character tags, aiType=2, animations), gray-list default avoidance, R-18 limited to official xRestrict, 1.5-year half-life favorite sorting, session-level deduplication, and a 15-second cooldown. It also states that files are written to local disk and metadata is returned, which is exactly the beyond-schema behavior 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?

A single dense paragraph, but the core purpose is front-loaded in the first sentence and subsequent clauses are rules rather than filler. The closing caption/display instruction is presentation guidance rather than invocation semantics, which slightly dilutes focus, but nearly every sentence earns its place given the tool's complexity.

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 5-parameter tool with no output schema and no annotations, the description covers filtering behavior, ordering, session semantics, rate limits, credential requirements for r18, and what is returned (local paths plus metadata, with caption format). An agent has everything needed to call it correctly and set user expectations.

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 real meaning: multiple tags are AND-ed against tags or titles, session isolates dedup history and cooldown, and allow_gray categories must genuinely be present or the item is rejected. It also warns that unmapped Chinese names should be converted to Pixiv-style Japanese tags, which the schema does not convey.

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 opening sentence states a specific verb + resource + full outcome chain: search Pixiv illustrations by character/tag, download originals locally, return file paths and metadata. The scope (character/tag-driven search plus download) is distinct enough that an agent can tell it is the discovery+fetch tool rather than a detail or raw-download tool.

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 gives concrete when-to-use rules for the risky options: allow_gray only when the user explicitly names a category, r18 only on explicit request, gray-list items must genuinely be present, and it advises pre-translating tags not in the built-in mapping. It stops short of explicitly contrasting with pixiv_download or pixiv_illust_detail, so sibling routing is left implicit.

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