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

Annotations already declare readOnly/idempotent/non-destructive, and the description still adds real behavior: README.md files are skipped, a package with no matches returns an empty list rather than an error, keys are sorted and uncapped, and every call rescans local files (i.e., no cached/stale view). These are non-obvious traits an agent cannot infer from the schema or 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?

Purpose and scan scope are front-loaded, followed by routing guidance, then parameter constraints and edge-case behavior. Every sentence carries distinct information; nothing is restated from the name, title, or schema.

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 explained, and annotations already carry the safety profile. The description still covers scan scope, parameter form, skip rules, empty-result behavior, ordering, and refresh semantics, leaving no gap an agent would need to guess at.

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 goes further by ruling out misuse ('exact directory name, not a glob, path, or regex'), stating the resulting scope (community/<package>/assets), and giving a concrete example value (校园情感领域包). It stops short of describing the multi-package precedence between community and 05_资产库 when omitted, but adds clear value over 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?

The description gives a specific verb and resource ('List asset files and the uppercase keys they declare') plus the exact scan roots (community/*/assets and 05_资产库/用户自定义), which no sibling shares. It even names the sibling reads it is not (asset_read, module_read, pattern_list), so the agent can distinguish it 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?

Explicit when-to-use ('discover a key before asset_read') and explicit when-not-to-use with named alternatives ('do not use it to fetch asset text (asset_read) or to list modules/patterns (module_read / pattern_list)'). This is a complete routing rule, not an implication.

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 well beyond the annotations: it discloses the default search scope (all community assets plus 05_资产库), that a package limits the scan to one directory with no fallback, that multiple matches may be returned, that no match is an error, and that full untruncated text is returned. Annotations only cover safety; the description carries the substantive 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 purpose, then constraints, then scope — good ordering with no filler sentences. The telegraphic style ('Tokens come from each file's name or first 6000 characters') is dense enough that a first-time reader may need a re-read, costing some clarity.

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 explained, and the description covers everything else an agent needs: valid input form, failure conditions, search scope, and multiplicity of results. Nothing material is missing for a two-parameter read tool.

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?

Although schema coverage is 100%, the description adds real semantics the schema lacks: keys are exact uppercase and never glob/regex/package/path/prose, lowercase or partial keys fail, and package narrows files but never key matching. It also explains that omitting package searches all community assets and 05_资产库.

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 — 'Read asset files for a key' — and even names the return shape (package, path, text). It explicitly distinguishes itself from asset_list, module_read, and pipeline_read, so an agent can route without opening sibling 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 explicit when-to-use ('Use only a key from asset_list') and when-to-use-something-else ('for enumeration use asset_list, for other resources module_read or pipeline_read'). The alternative routing 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?

Annotations already declare readOnly, idempotent, non-destructive, and closed-world, so the safety profile is covered. Beyond that, the description adds non-obvious behavior: filtered sources are omitted without reordering the remainder, count is post-filter, and it 'reads only the declaration, never source bodies' — meaningful operational context not derivable from annotations or 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?

Front-loaded with the return contract, then usage, then parameter semantics, then a closing read-scope caveat. Three sentences with no filler; every clause either routes the agent, defines the parameter, or discloses filtering behavior.

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-shape explanation is not required, and the description correctly focuses on when to call, when not to, and how clearance filtering behaves. For a zero-required-parameter read tool with an output schema, nothing an agent needs to invoke 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 enum values are already documented, so the baseline is 3. The description goes further by explaining the ceiling semantics concretely — 'public keeps public sources, internal keeps public and internal, restricted keeps all' — and clarifying the omit behavior, which adds interpretive value beyond the enum list.

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 opens with a specific verb+resource pair ('Return the declared knowledge-source query order and a contract-first flag'), naming exactly what is produced. It then explicitly distinguishes itself from siblings by naming library_search, module_read, and asset_read as the tools for the jobs this one does not do.

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 trigger ('Use it before assembling or citing NF content') and two explicit exclusions with the correct alternatives ('do not use it to search documents (use library_search instead)', 'read module/asset bodies (module_read / asset_read)'). It also advises 'use public before republishing material', a concrete usage condition for the single parameter.

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, idempotentHint=true, and openWorldHint=false. The description adds valuable behavioral details beyond those annotations: unknown IDs return an error with the requested value and repair guidance, reads are local-only, the full text is returned without truncation, and there are no auth requirements or rate limits. This is rich transparency for a read-only 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 front-loaded with the core purpose, then immediately covers routing, ID format, error behavior, read characteristics, and environment constraints. Every sentence carries useful information, with no redundant or filler content.

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?

Given the tool's simplicity, the 100% schema description coverage, and the presence of an output schema, the description is fully complete. It covers purpose, sibling differentiation, ID format, error handling, return behavior, and local-only constraints, leaving no critical gap for an agent.

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 single parameter's case-insensitivity and example are already documented. The description adds additional format semantics not present in the schema: the ID is 'the canonical stem without the .md suffix' and matching is case-insensitive. This meaningfully extends parameter understanding beyond the structured field.

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: 'Read one library catalog entry by ID, returning its frontmatter record and body.' It also explicitly distinguishes this tool from siblings, naming library_search, module_read, pipeline_read, and asset_read as alternatives. An agent can select this tool without opening another 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?

Usage is explicit: 'Use it after library_search resolves the ID; do not use it to search ... or to read modules, pipelines, or assets.' This provides both when-to-use and when-not-to-use guidance, along with named alternatives. Nothing 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.

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

Annotations already declare readOnly, idempotent, non-destructive, closed-world, so safety is covered; the description adds substantial behavior beyond that: case-insensitive relevance-ranked catalog matching vs case-sensitive substring doc matching, catalog-first ordering that can hide doc hits, and the empty-query / no-match behaviors.

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 paragraphs: purpose and routing first, then edge-case behavior. Every sentence carries distinct information (matching rules, cap semantics, ordering hazard, empty-query result) with 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, return values need not be enumerated, and nothing an agent needs to call this correctly is missing: routing, matching semantics, cap behavior, and pagination absence are all covered.

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, but the description adds real semantics the schema lacks: the 1–50 cap is shared between catalog and document hits, and there is deliberately no offset/next-page parameter. This meaningfully clarifies query and limit behavior beyond the field descriptions.

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 (search) plus the two resources searched (library catalog and repository docs), and explicitly names the sibling tools it is not (library_read is the follow-up consumer, registry_search/spec_list cover exact metadata). An agent can route without opening another 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?

Explicit when-to-use (find an entry ID for library_read, locate a doc or package README) and when-not (exact module or protocol metadata, pointing at registry_search / spec_list). Nothing 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.

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

Beyond the readOnly/idempotent/closed-world annotations, it discloses the error behavior for unknown IDs (returns the requested value plus repair guidance), that metadata is resolved from a single local file, that the contract is not evaluated, and that there is no auth, network, rate limit, or write side effect. This is substantive behavior an agent cannot get from annotations alone.

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 action, then usage, then ID semantics, then behavior; every sentence carries information. Minor overlap between the trailing 'no auth, network, rate limits, or writes' and the existing annotations, but not wasteful enough to penalize more.

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 enumerated, and the description still clarifies what comes back (full file plus resolved metadata) and what is deliberately not done (contract evaluation). For a single-required-param read tool, nothing an agent needs 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 already 100%, so baseline is 3, but the description adds real meaning: the ID is a declared machine-contract/heading ID (M90) or a qualified ID (通用:M10) as returned by registry_search, explicitly not the filename, and matching is case-sensitive. It complements rather than repeats 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?

Names a specific verb (read), a specific resource (one NF module's full Markdown by ID), and the extra artifact returned (raw machine-contract block). It also explicitly distinguishes itself from pipeline_read and asset_read, so an agent can separate it from siblings without opening 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 the positive trigger (after registry_search resolves a module, or enumerate IDs via registry_search with an empty query) and the negative trigger (not for pipelines or assets, with the correct alternatives named). Both when-to-use and when-not-to-use are 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/idempotent/non-destructive, and the description adds genuinely new behavioral facts: metadata-only returns, no pagination, empty list on no match, and a fresh rescan of the local patterns directory on every call. The last point is a real cost/consistency signal not derivable from 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?

Front-loads the core purpose and return shape, then routes to alternatives, then the operational caveats. Dense but each clause carries distinct information; slightly long for a zero-parameter list tool.

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, yet the description still summarizes the return shape and edge cases (empty list, no pagination, metadata only). Combined with the strong routing guidance, an agent has everything needed to call and interpret it.

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?

The tool takes zero parameters, so the baseline is 4; the description correctly implies no filtering or paging inputs, but with an empty schema there is nothing further to add.

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 ('List engineering patterns as id/path/title/status metadata from patterns/*/PATTERN.md, sorted by ID') and explicitly contrasts itself with pattern_read, module_read, and asset_list, so an agent can distinguish it from every sibling without opening a 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?

Gives both the positive case ('Use it to discover a pattern ID before pattern_read') and the negative case ('do not use it to fetch pattern text ... or to list modules/assets'), naming the correct alternatives for each. Nothing about when to select this tool 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.

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

Beyond the readOnly/idempotent/closed-world annotations, it discloses error semantics ('a missing pattern is an error, not an empty result'), output fidelity ('full text without truncation'), and operational context ('Local-only; no auth or rate limits'). These are non-obvious traits the agent needs and are not in 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?

Front-loads purpose then routes alternatives, parameters, and error behavior. Dense but every clause carries information; slightly packed into long sentences rather than cleanly separated.

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 format needn't be restated, and the description still notes full untruncated text. For a single-file read tool, an agent has everything needed to call it correctly and interpret errors.

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, but the description adds real meaning: pattern_id is case-sensitive and is the directory name, not the document title (e.g. fail-closed-verification). That disambiguation goes beyond the schema's own wording.

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 ('Read one engineering pattern's full Markdown') plus the exact location (patterns/<id>/PATTERN.md) and contents (frontmatter, rules, counterexample). It also names the sibling tools it is not, so an agent can distinguish it 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?

Gives an explicit use case ('for the practice behind an NF rule') and explicit exclusions with named alternatives: pattern_list for enumeration and module_read for domain modules. Nothing is left to inference about when to pick this tool.

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 read-only/idempotent/non-destructive, but the description adds substantive context beyond them: each call rescans the repository, results are sorted by repository path, no matches yields an empty list, and it is local-only with no auth or rate limits. It also reiterates that bodies are never returned, which is the key behavioral boundary.

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 lead sentence front-loads scope, the second routes usage, and the remainder packs only actionable detail with zero filler. Dense but every clause carries information an agent needs.

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 one optional parameter, full schema coverage, an output schema, and annotations, the description closes the remaining gaps (rescan cost, sort order, empty-result behavior, no auth). Nothing needed to invoke 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%, so baseline would be 3; the description earns extra by clarifying that the query matches the repository-relative path or H1 title and not file content, that wildcard characters are literal, and by giving concrete example values (P04, 技术文档, 校园). It stops short of 5 only because the core matching rule is already in 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 and resource (list NF pipelines), names the exact locations scanned (03_管线库 and community/*/pipelines), and clarifies it returns metadata only. An agent can distinguish it from pipeline_read, registry_search, and spec_list without opening the 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?

Explicitly prescribes when to use it (discover a pipeline ID before pipeline_read) and enumerates three do-not-use cases with the correct alternative named for each. No inference required.

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/closed-world, and the description adds context beyond them: local-only with no auth or rate limits, a missing pipeline is an error rather than an empty result, no truncation, and path-validation rules rejecting roots outside 03_管线库/ and community/*/pipelines/ and '..' escapes.

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 action, then constraints and error semantics in tight order; content is dense but every sentence carries a distinct fact. Slightly longer than minimal, but 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?

For a single-parameter read tool with an output schema, everything an agent needs is present: usage routing, accepted identifier forms, path restrictions, and error behavior. Return format is correctly left to the output schema.

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 schema already documents the single 'pipeline' parameter, but the prose adds real meaning beyond it: the Pxx stem is matched against the filename prefix, and the accepted path roots and the '..' escape rejection clarify the accepted syntax.

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: 'Read one NF pipeline's full Markdown by P-number or repository-relative path.' It explicitly distinguishes itself from pipeline_list, module_read and asset_read in the same breath, so an agent can select it 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?

Gives explicit when-to-use ('after pipeline_list identifies a pipeline') and when-not-to-use with named alternatives for both the listing case (pipeline_list) and the module/asset case (module_read / asset_read). Routing guidance is complete.

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/idempotent/non-destructive, so the bar is lower, yet the description adds substantial context: reads registry.json only, never scans module or pipeline files, never matches descriptions/paths/file contents, no matches yields empty arrays, and results are grouped with protocol hits secondary. This is disclosure well 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?

Front-loaded with purpose, then filtering behavior, then exclusions and edge cases in one dense block. Every clause earns its place, though the long semicolon-chained sentences are heavier than ideal for quick scanning.

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 single-required-param read tool with an output schema, the description covers matching semantics, edge cases, exclusions, and result shape (grouped modules/protocols). Nothing an agent needs to invoke it correctly is missing, and return-value detail is appropriately left to the output schema.

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 schema already states case-sensitive substring semantics, so baseline would be 3. The description adds value beyond the schema by clarifying required-ness, that an empty query matches every entry, and by giving concrete examples (M90, 技术文档域包).

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 (query/search) and resource (NF registry for modules), with explicit scope: id or name contains query. It names the siblings it is not (spec_list for the protocol catalog, library_search for free-text documents), so an agent can route correctly without opening 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 an explicit when-to-use (resolve a module ID before module_read) plus two named exclusions with their intended alternatives. This is the full when/when-not/alternatives pattern.

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 the safety profile (readOnly, idempotent, non-destructive, closed-world), yet the description still adds genuine behavioral context: reads one local file deterministically, missing/malformed registry is an error, empty list means no protocols, and results follow registry order. That is meaningful disclosure beyond 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?

Purpose and scope are front-loaded in the first sentence, and most sentences earn their place. The closing warning about registry_search protocol hits restates the earlier exclusion and costs some density, keeping it just short of ideal.

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 no explanation, but the description still covers the edge cases an agent needs: no-arg invocation, empty-result meaning, error on missing/malformed registry, and deterministic ordering. Complete for a zero-param read tool.

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?

Zero parameters, so the baseline is 4; the description confirms 'Takes no arguments', consistent with the empty schema. Nothing further is needed since there is no parameter surface to clarify.

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 ('List protocol packages') and even names the source of truth (desktop/src/core/registry.json) plus the exact fields returned (id, version, module count, categories). It explicitly contrasts itself with registry_search and pipeline_list, so an agent can distinguish it from all listed 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?

Explicit when-to-use ('authoritative protocol package catalog') and when-not-to-use, naming both alternatives (registry_search for modules, pipeline_list for pipelines). It even preempts a likely mistake by warning that protocol hits from registry_search are not a substitute.

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