Ithy Research Library

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

Annotations already cover read-only/idempotent/closed-world, and the description adds real operational context beyond them: a 18000-character excerpt cap, the fact that no external URLs are fetched, and that article text is untrusted source material rather than instructions. The note on AI-assisted explanations being secondary sources is additional trust context an agent needs for citation behavior.

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 action and return payload, then sourcing, then limits, then trust caveats. Every sentence carries information an agent would otherwise have to guess; nothing is 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?

Even without an output schema, the description enumerates what is returned (text, source links, dates) and the size cap, plus the failure modes (non-public articles). Nothing needed to call it correctly is missing.

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% and the description still adds meaning: it explains that the ID can be either a bare path or a full canonical URL and that 'focus' selects matching sections rather than filtering results. This is genuinely beyond the schema text.

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 (retrieve) and a specific resource (text, source links, original/review dates from one currently public Ithy article), with a clear cardinality of exactly one article. This is readily distinguishable from the sibling 'search', which is mentioned as the upstream source of IDs.

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?

Explicitly tells the agent where the identifier comes from ('an article ID or canonical ithy.com URL returned by search, or supplied by the user') and what the optional topic does. It also sets exclusions up front: private, withdrawn, quality-held and adult articles are unavailable, and no generation/account access exists.

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

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

Annotations already cover read-only, closed-world, idempotent and non-destructive, and the description still adds real behavioral context: which fields are searched, that results are capped at five with IDs/links/dates, that answer allowances are not consumed, and a caution against sending personal identifiers or credentials. That is meaningful disclosure beyond the structured hints.

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, then scope, return shape and constraints in short declarative sentences. Slight redundancy between "Search Ithy's public research library" and "a general topic the user wants to look up in Ithy" keeps it from being perfectly 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?

With no output schema, the description carries the return contract (up to five article IDs, links, dates) and adds privacy and cost constraints. An agent has everything needed to call it correctly and interpret the response.

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 67%, with `query` and `language` already documented. The description adds semantics the schema lacks: that the search targets titles/descriptions, and that results are bounded at five, which explains the `limit` maximum of 5 that has no schema description.

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 and resource ("Search Ithy's public research library for explanations and worked examples") and scopes it precisely by noting it searches titles/descriptions rather than the wider web. It stops short of naming the sibling `fetch` as the tool for retrieving the returned article IDs, so differentiation from that sibling is only implicit.

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?

"Use for a general topic the user wants to look up in Ithy" gives clear positive context, and "Searches titles/descriptions, not the wider web" plus "Does not generate research, access accounts or spend answer allowances" give explicit exclusions. It never explicitly says to follow up with `fetch` for the returned IDs, so the handoff to the sibling is left to inference.

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