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

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered; the description adds real behavioral value by disclosing that page ranges, DOIs and names are normalized during conversion and that source format is auto-detected. It does not state failure behavior on malformed input, leaving a modest gap.

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?

Compact and front-loaded, with formats named first and the detection caveat last. The example sentence is somewhat awkwardly phrased and restates the obvious mapping, but it does illustrate the required parameters, so it 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?

With an output schema present the description needn't explain return values, and parameters are fully documented in the schema; the added normalization note covers the main hidden behavior. Only the plain-text-as-input ambiguity and error handling are unaddressed, which is minor for a stateless read-only converter.

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 both enums are self-documenting, so the baseline is 3; the description earns the extra point by clarifying that 'from' defaults to detection and that specifying it suppresses auto-detection. Note it describes plain text as a conversion target but the 'from' enum omits it, leaving input-side text handling slightly ambiguous.

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 (convert) and resource (a reference list) plus the exact formats supported, which lets an agent distinguish it from parse_bibliography, dedupe_entries, lint_entries and generate_cite_keys without opening any 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?

Explains the default behavior (input format is auto-detected) and the one override case ('Set from only to override detection'), giving clear context for invocation. It stops short of explicitly routing the agent away from sibling tools such as parse_bibliography, so a 4 rather than a 5.

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 and openWorldHint=false, and the description reinforces that with 'Your files are not changed.' It goes further by disclosing the output shape – groups with a keep/remove decision and a deduplicated BibTeX list – plus the keep-the-entry-with-most-fields tie-break rule, which is useful behavioral context beyond the annotations.

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 matching rules and non-mutation guarantee are front-loaded, and the worked example clarifies the tie-break rule. The example restates what the preceding sentences already say, so it is slightly redundant, but overall the text is efficient.

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 needn't be explained, and the description still summarizes the group structure and BibTeX output. It covers matching rules, tie-breaking, and non-mutation; only minor gaps remain, such as how format auto-detection behaves or when to adjust the title threshold.

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 the schema already documents text, format, and title_threshold. The description alludes to title similarity matching but adds no syntax, threshold guidance, or format-detection detail beyond what the schema provides, 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 ('Find duplicate references') and enumerates the exact matching rules: DOI, shared cite key, and similar titles with non-clashing years. This is clearly distinguishable from siblings like lint_entries or parse_bibliography, which do not report duplicate groups.

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 context of use is implied by the matching criteria, so an agent can infer this is the tool for spotting duplicate entries. However, it never states when to prefer it over lint_entries or parse_bibliography, nor any exclusions or prerequisites.

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 and openWorldHint=false, so the description need only add context — and it does: repeated keys get a letter suffix, and the response contains the old/new key pair per entry plus a full BibTeX list. It doesn't mention any rate limits or failure modes for unparseable input, keeping it short of a 5.

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 rule, then collision handling, then return contents — a sensible order. The trailing example restates the smith2024deep pattern already given and is longer than strictly necessary, so it loses a point for redundancy rather than length.

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 needn't be spelled out, yet the description covers the old/new key mapping and BibTeX output anyway. The only gap is guidance on which input formats are accepted beyond the schema's enum and what happens with malformed text, but for a 3-param read-only tool this is close to 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 coverage is 100%, so baseline is 3. The description goes beyond the schema by demonstrating include_title_word's effect concretely (smith2024deep vs smith2024) and by implying format detection behavior, though it never discusses the format parameter's auto/bibtex/ris/csl_json choices.

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 (create cite keys) plus the exact derivation rule (first author's family name, year, first meaningful title word) and the collision behavior. Nothing about parse_bibliography, convert_bibliography, dedupe_entries or lint_entries overlaps with this key-generation transformation, so the agent can distinguish it immediately.

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 description makes clear what the tool produces but never says when to choose it over siblings such as convert_bibliography or dedupe_entries, nor does it state prerequisites (e.g. the text must already be parsed/unambiguous). Usage is only implied by the nature of the operation.

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 establish readOnlyHint=true and openWorldHint=false, and the description adds substantial behavioral detail beyond that: it names the issue categories, explains that each issue reports the entry key, field, and message, and gives a concrete missing-field example. This tells the agent both what is checked and the shape of the findings.

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 sentences, front-loaded with the core purpose, followed by output semantics and a compact example. Every sentence contributes useful information and nothing is redundant.

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 linting tool with two well-documented parameters and an output schema, the description supplies everything needed: the scope of checks, the issue-reporting shape, and the key usage constraint about passing the full list.

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 meaning by explaining why the entire list should be passed as text: repeated-key detection requires the full document. The format parameter is not discussed, but its schema description and enum already make its behavior clear.

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 (check/lint) and resource (reference list) and enumerates the exact problem classes it detects. An agent can distinguish it from parse_bibliography, convert_bibliography, dedupe_entries, and generate_cite_keys because those do not diagnose field, page, capitalization, DOI, year, or key issues.

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 description makes clear that this is the diagnostic tool for a reference list and adds the practical instruction to pass the whole list as text so repeated keys can be found. It does not explicitly name when to choose it over each sibling, but the intended context is unambiguous.

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?

Goes beyond the readOnly/openWorld annotations by disclosing that processing is local ('Nothing is sent anywhere') and by spelling out the exact normalization transformations applied to fields, DOIs, years, and author names. It doesn't say how malformed or unparseable input is handled, which is the remaining behavioral gap.

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 opening sentence front-loads the core purpose and normalization behavior; the example earns its space by showing input and expected authors/field casing. It's slightly dense but contains no 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 an output schema present, full parameter coverage, and annotations covering safety, the description supplies the remaining needed context: accepted formats, normalization contract, and a concrete example. Minor gaps remain around error behavior for invalid or mixed-format input.

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, making 3 the baseline. The description adds practical meaning by advising to omit format unless auto-detection fails, and the worked example shows what the text parameter looks like in practice.

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 clear verb and resource: read BibTeX/RIS/CSL JSON text and return entries in one normalized shape. The normalization rules (lower case fields, bare DOIs, year from date, authors split) make the operation concrete. It doesn't explicitly distinguish itself from sibling convert_bibliography, which is the nearest overlap, so it stops short of a 5.

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?

Provides clear guidance for the one ambiguous parameter: 'Leave format out unless auto picks the wrong one,' which tells the agent to rely on auto-detection first. Usage context is clear from the opening sentence, but there is no explicit statement about when to use this versus convert_bibliography or the other siblings.

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