SourceToMe

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

Annotations already declare readOnlyHint=true and destructiveHint=false, so safety is covered. The description adds meaningful non-obvious behavior: it recomputes selection, discloses the deduplication rule ('Edition duplicates cannot inflate independent support'), and clarifies the output's semantic boundary. It stops short of describing the actual return shape despite having output schema, which is fine.

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 tightly packed sentences, front-loaded with the core action, then the deduplication rule, then the scope caveat. No filler; every clause adds value.

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 read-only analysis tool with an output schema, the description covers purpose, key heuristic behavior, and scope boundaries well. The only omission is parameter guidance for source_ids, but the output schema reduces the need to explain return values, so the definition is largely complete.

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 0% – neither 'question' nor 'source_ids' has a description in the schema. The description mentions the concept of a question but provides no guidance on source_ids (optional array, max 4, specific ID pattern) or its role in scoping the selection. With the schema silent and the description not compensating, parameter meaning is a real gap.

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 ('Recompute evidence selection... and explain... heuristic coverage') and names the specific output categories (Strong, Limited, Insufficient). The closing clause 'This explains selection, not factual correctness' decisively distinguishes it from siblings like find_evidence_gaps and retrieve_evidence.

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 phrase 'This explains selection, not factual correctness' sets a clear scope boundary about what the tool is for and implies what it is not. However, it does not name specific sibling alternatives (retrieve_evidence, find_evidence_gaps) or state when to choose this over them, leaving some routing inference to the agent.

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=true, openWorldHint=false and destructiveHint=false, and the description reinforces this by promising no retrieval/acquisition and no generated answers. It adds genuinely useful output-shape context: results are gap suggestions, not evidence, which sets expectations the annotations cannot convey. Auth, rate-limit and latency behavior remain unstated.

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?

Two tight sentences, purpose front-loaded, with the scope exclusions compressed into one semicolon clause. Nothing is redundant and nothing needs to be re-read.

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?

An output schema exists, so return-value description is rightfully omitted, and the behavioral boundaries are covered well enough to call the tool safely. The single material hole is the undocumented 'source_ids' parameter, which an agent needs to decide whether to pass.

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 0% and the description never mentions the parameters. The 'question' argument is self-evident, but 'source_ids' (bounded to 4 ids matching ^public-pg-[0-9]+$) is completely unexplained — an agent cannot tell whether it scopes the gap search, supplies known sources, or seeds the analysis.

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 names a specific verb and resource: it identifies corpus limitations and the kinds of evidence a question would need, specifically when the question is unsupported. The closing clause ('suggestions are gaps, never retrieved evidence') draws a hard line against the retrieval siblings (retrieve_evidence, search_library), so an agent can route without opening either schema.

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 states the triggering condition clearly — use this for an unsupported question where evidence is lacking — and enumerates what it will not do (no web search, acquisition, factual answer, personalized advice). It never names a sibling tool as the alternative for those excluded cases, so the routing is implied rather than explicit.

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 readOnly/not-destructive/closed-world, so the safety profile is covered. The description adds real value past that: an access-scope constraint ('public' sources only, no private or arbitrary ones), which the annotations do not express.

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?

Two sentences, no filler. The scope and the enumeration of inspectable content come first, the limitation second; every clause earns its place.

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?

An output schema exists, so return values need not be spelled out, and annotations cover the safety profile. The description supplies the remaining piece an agent needs – source-ID provenance and the public-only restriction – leaving no material gap.

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 0%, so the description carries the burden for the single source_id parameter. It compensates reasonably by stating the ID must be a public ID obtained from search, matching the '^public-pg-[0-9]+$' pattern, though it does not restate the ID's format explicitly.

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 ('Inspect') and resource ('one public source ID'), and enumerates what is returned (work, edition, translator, catalog, fingerprint, locator inventory). The phrase 'returned by search' positions it downstream of search_library and separates it from the evidence-oriented siblings.

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?

Implies the prerequisite (an ID produced by search) and gives an explicit exclusion: private or arbitrary sources cannot be inspected. It stops short of naming a sibling alternative or stating what to do when the ID is not from search, so it is clear context rather than full when/when-not 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=true, destructiveHint=false and a closed (openWorldHint=false) corpus, so the safety profile is covered. The description adds two genuine facts beyond that — the scope is a 'bounded public library' and missing evidence is reported rather than erroring — but nothing about volume or ranking 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?

Three dense sentences with the core purpose front-loaded and the two usage modes immediately after. Every clause carries information, though the locator sentence is packed tightly enough to need a second read.

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 an output schema present, return values need not be described, and the description covers purpose, the two invocation modes and the missing-evidence case. For a 4-parameter, 0%-coverage schema the undocumented 'question' and 'limit' parameters are the one remaining gap.

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 0%, so the description carries the full burden, and it supplies the non-obvious cross-parameter rule that a locator must be paired with exactly one source ID, plus the semantic effect of multiple source_ids. It leaves 'question' and 'limit' (default 5, max 5) entirely unexplained.

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 concrete verb+resource pair ('Return passages, edition, locator and provenance') and names the corpus it draws from ('bounded public library'). It is distinguishable from search_library and inspect_source in spirit, but never names or contrasts a sibling explicitly.

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 real conditional guidance: one source ID plus a full locator retrieves an exact indexed locator, while multiple IDs switch the tool into edition-comparison mode. It stops short of stating when-not-to-use-this or naming search_library as the alternative for open-ended discovery.

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 readOnly, non-destructive and closed-world, so the safety profile is covered. Beyond that, the description adds genuinely useful behavior: the corpus is a fixed set of four sources, and stable IDs are returned 'even when no passage matches' for follow-up — non-obvious behavior an agent could not get from the schema.

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?

Two sentences, no filler, with the corpus identity and scope front-loaded. The second sentence is semicolon-heavy and slightly overloaded with the ID-return and exclusion details, 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?

An output schema exists so return values need not be described, and annotations cover the safety profile. With corpus, scope exclusions and follow-up ID behavior all stated, an agent has what it needs to call this 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 description coverage is 0%: query, limit and source_ids carry no documentation. The description hints that the four named sources are addressable and that IDs are stable, which partially illuminates source_ids, but the query semantics (min 3 / max 600 chars) and the limit cap of 5 are left entirely unexplained.

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 names a specific verb ('Find passages') and resource ('public reference library'), then enumerates the exact holdings (Pride and Prejudice, Frankenstein 1818/1831, The Art of War). It scopes the tool with 'no personal library or internet search,' but never distinguishes it from named siblings like retrieve_evidence or inspect_source, so the agent must infer the boundary.

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?

'No personal library or internet search' tells the agent when not to reach for this tool, which is real guidance. However, no alternative tool is named for those cases, and there is no statement of when to prefer this over the sibling search-like tools, leaving usage only partially implied.

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