OKA Anywhere

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

Adds substantial detail beyond the readOnly/idempotent/closedWorld annotations: the return format (written answer plus a Sources list of verbatim quotes checked against documents), explicit refusal behavior when documents don't cover the question, and disclosure that an LLM generates the answer. These are exactly the traits an agent needs and none are in structured fields.

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?

Four sentences, each doing distinct work: purpose, when to use, what is returned, and failure behavior, with the core purpose front-loaded. No repetition 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?

No output schema exists, and the description compensates by describing the answer-plus-Sources return shape and the not-covered response. Combined with full schema coverage, an agent has everything needed to call it and interpret results.

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 100%, so both question and optional corpus are documented in the schema itself. The description adds no further parameter detail and does not mention the corpus override at all, so the baseline 3 applies.

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: answering a question from the office's approved procedure documents. The scope (only approved procedures, plain-language questions) distinguishes it from siblings like search_procedures or verify_claim without needing their schemas.

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 clear usage conditions: any question about how the office does something, such as a policy, a step, or who to ask. It also warns that rephrasing a question won't change a not-covered answer, which prevents retry loops. It stops short of naming when to prefer search_procedures or verify_claim instead.

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 readOnly, idempotent, non-destructive, and closed-world behavior. The description adds genuinely new traits: the return shape (boundary statements grouped by document title, or a note that none were found) and the fact that it takes no question and uses no language model, signaling deterministic, non-hallucinated output versus the LLM-backed siblings.

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?

Three tight sentences, front-loaded with the purpose, then usage, then return behavior and the no-LLM caveat. Every sentence carries distinct information with no repetition of the schema or annotations.

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 compensates by describing the return grouping and the empty-result case, plus the deterministic no-model behavior. For a zero-required-parameter, read-only listing tool, an agent has everything needed to decide and call 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 description coverage is 100%, and the single optional corpus parameter is fully documented in the schema (including the default and refusal of made-up ids), so the baseline is 3. The description's 'takes no question' clarifies the absence of a query parameter, a mild addition given the question-taking siblings, but adds no corpus semantics 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?

States a specific verb+resource (list boundary statements from approved documentation) and defines the scope precisely as what is 'explicitly forbids or says it does not cover.' This is clearly distinguishable from the sibling tools ask_procedures, search_procedures, and verify_claim, which retrieve or check content rather than enumerate documented gaps.

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 explicit triggers: use it before concluding something is allowed, or when a question is not covered, to check whether the topic is a documented gap. It establishes context well but never names an alternative sibling (e.g., verify_claim or ask_procedures) or states when this tool is the wrong choice, so it stops short of full routing guidance.

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 declare readOnlyHint, idempotentHint, openWorldHint=false, and destructiveHint=false. The description adds substantial behavior beyond that: it avoids an LLM, returns up to three titles with excerpts, may return no matching procedure, and ranks by keyword overlap. These details directly shape how an agent should interpret and invoke the tool.

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 compact and front-loaded, opening with the core action and distinguishing clause. Each sentence carries distinct information: purpose, usage, return shape, and ranking behavior. There is no repetition 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?

With no output schema, the description explains the return format in prose: up to three document titles with excerpts, or a no-match result. Combined with complete parameter descriptions and annotations covering safety traits, the description supplies what an agent needs to call the tool 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. The description still adds query-construction guidance beyond the schema: ranking is by keyword overlap, so the agent should use words the documents are likely to use. It does not add further corpus semantics, but the query guidance is genuinely useful.

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 states a specific verb and resource: find relevant passages in approved procedure documents. It also distinguishes this from an LLM-based sibling by saying 'without a language model' and by describing a keyword-overlap ranking. The scope and return shape are clear enough to separate it from ask_procedures.

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 clear usage contexts: read source text, see what exists on a topic, or when an exact sentence is not found. It also gives practical query guidance ('use the words the documents are likely to use'). However, it does not explicitly name ask_procedures or state when to use that alternative instead.

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 cover the safety profile (readOnly, idempotent, non-destructive, closed-world), and the description goes well beyond them: it specifies the return contract (VERIFIED plus document title and surrounding quote), the strict meaning of VERIFIED (whole statement found together in one document), partial-match handling, normalization rules for case/spacing/curly quotes, and that no language model is used.

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-loads the purpose and then layers return behavior and matching rules in short, information-dense sentences. There is mild overlap between the 'returns VERIFIED ... or not found verbatim' sentence and the following explanation of partial matches, but nothing is wasted.

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 full return-value burden and does so: it explains the VERIFIED/not-found outcomes, what accompanies each, and the exact-match boundary conditions. An agent has everything needed to call it and interpret the result 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 both parameters are already documented in the schema, including the 'leave it out for the default corpus' guidance on corpus. The description adds only the general verbatim-matching semantics, not parameter-level syntax or format detail, so the baseline of 3 applies.

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: checking whether an exact statement appears word for word in the office's approved procedure documents. The verbatim requirement plus the 'no language model' note clearly separates it from a semantic sibling like ask_procedures, though the description never names the sibling tools it differs from.

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 an explicit use case ('confirm a specific sentence, number or rule before relying on it') and an implicit exclusion via 'a paraphrase will not match', which steers paraphrased lookups elsewhere. It stops short of naming search_procedures or ask_procedures as the alternative for fuzzy or conceptual queries.

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