Server Quality Checklist

Tool Scores

• importPlaylistC
  2.4/5.0
  Behavior2/5

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

  The timing bracket provides useful behavioral context not found in annotations, but critically fails to disclose mutation semantics: the 'reindexExisting' and 'activateCollection' parameters suggest this overwrites data and modifies system state, which the description omits entirely.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness3/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The single sentence is front-loaded with the core purpose and avoids redundancy, earning its place. However, it is inappropriately brief for a 14-parameter configuration tool, lacking the necessary depth despite being structurally efficient.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness1/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Grossly inadequate for the tool's complexity. With 14 parameters (including enums, ranges, and boolean flags), no output schema, and no annotations, the description provides insufficient context for safe and effective invocation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters1/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage across 14 parameters, the description completely fails to compensate. No parameter semantics are provided for critical configuration options like chunkStrategy, embeddingProvider, or the distinction between collectionId and label.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the specific action (import), resource (playlist), destination (local transcript knowledge base), and purpose (semantic search). However, it does not distinguish from sibling tool 'importVideos' or clarify when to prefer this over analyzing playlists without importing.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The timing annotation [~5-30s, ~2s per video] provides implicit guidance about scale expectations, but there is no explicit guidance on when to use this versus 'importVideos' for individual videos, or prerequisites like collection setup.

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

• expandPlaylistC
  2.6/5.0
  Behavior2/5

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

  With no annotations provided, the description carries the full burden. It adds valuable timing context ('[~1-3s]') but fails to disclose critical behavioral traits suggested by the 'dryRun' parameter (implying side effects), nor does it explain return values, idempotency, or data persistence.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is appropriately brief and front-loaded with the core action. The timing notation, while slightly informal, efficiently communicates performance expectations without verbosity. No redundant or filler content is present.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given four parameters with zero schema descriptions, no output schema, and no annotations, the description is insufficiently complete. It omits return structure, parameter semantics, and side-effect disclosures that are essential for a tool with a 'dryRun' option suggesting mutating behavior.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters1/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, requiring the description to compensate for undocumented parameters. However, the description fails to explain any of the four parameters (playlistUrlOrId, maxVideos, includeVideoMeta, dryRun), leaving critical constraints like the 200-video limit and the purpose of dryRun completely undocumented.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly defines the transformation ('Expand a playlist into individual videos') and identifies the target use case ('downstream analysis and batch workflows'). It effectively distinguishes from siblings like analyzePlaylist and importPlaylist by emphasizing the expansion/decomposition aspect, though it doesn't explicitly contrast with them.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description provides implied usage context ('for downstream analysis') but lacks explicit guidance on when to use this tool versus alternatives like importPlaylist or analyzeVideoSet. No 'when-not' scenarios or prerequisites are mentioned.

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

• compareShortsVsLongC
  2.7/5.0
  Behavior2/5

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

  No annotations provided. The description mentions output ('suggest a format mix') and includes a cryptic latency hint '[~3-10s]', but critically fails to clarify the presence of 'dryRun' parameter — leaving ambiguity whether this tool performs read-only analysis or modifies channel settings when dryRun=false.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness3/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single sentence is appropriately concise and front-loaded, but the bracketed notation '[~3-10s]' is structurally cryptic and unexplained (execution time estimate? Shorts duration reference?), creating confusion rather than clarity.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a 3-parameter tool with zero schema documentation and no output schema, the description is incomplete. It lacks parameter documentation, output format details, and behavioral constraints necessary for safe invocation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters1/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 0% description coverage. Description completely fails to compensate: 'channelIdOrHandleOrUrl' accepts multiple formats (ID/handle/URL) not explained; 'lookbackDays' range implications unstated; 'dryRun' purpose (preventing modifications?) is undocumented despite being critical for safe invocation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description clearly states the tool 'Compare[s] recent Shorts vs long-form performance for a channel and suggest[s] a format mix' — specific verb (compare), specific resource (Shorts vs long-form performance), and distinct scope from siblings like analyzeVideoSet or inspectChannel.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No guidance provided on when to use this versus alternatives like analyzeVideoSet or inspectChannel. No mention of prerequisites (e.g., channel authorization) or when the format mix suggestion is applicable.

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

• listCommentCollectionsC
  2.7/5.0
  Behavior3/5

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

  Without annotations, the description carries the full disclosure burden. It provides one behavioral trait (~instant speed) and summarizes the returned data scope, but lacks details on output format, pagination, or side effects that would help an agent interpret results.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness3/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  While the single-sentence structure is efficient and front-loaded, it is inappropriately terse given the complete lack of schema documentation for the parameter. The bracketed [~instant] notation is concise but cryptic.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given zero schema coverage, no annotations, and no output schema, the description should explain the parameter and output structure. It only partially fulfills this by listing conceptual return values while leaving the tool's configuration and exact return format undocumented.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters1/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage for the includeVideoList boolean parameter, the description completely fails to compensate. It makes no mention of this parameter or how it modifies the output (presumably toggling video list inclusion in the response).

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses a specific verb (List) and clearly identifies the resource (comment collections), distinguishing it from the sibling listCollections tool. It also specifies the three elements returned (collections, active search focus, and indexed counts), providing concrete scope.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description provides only a performance hint ([~instant]) but offers no guidance on when to use this versus alternatives like searchComments or readComments, nor does it explain prerequisites or relationships to the 'active collection' concept implied by sibling tools.

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

• scoreHookPatternsC
  2.7/5.0
  Behavior2/5

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

  With no annotations, the description carries full burden but omits critical behavioral details: whether results are persisted (implied by dryRun parameter but not explained), what the score format/range is, or whether this requires specific permissions. Only provides latency estimates ('~3-10s').

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise with only two components: functional description and performance estimate. Front-loaded with the action. The bracketed timing notation is slightly cryptic but efficient. No wasted words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given 0% schema coverage, no annotations, and no output schema, the description leaves significant gaps: parameter semantics are largely undocumented, return value structure is unknown, and side effects (suggested by dryRun) are unexplained. Needs expansion for a 3-parameter mutation-capable tool.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 0% description coverage. The description mentions 'first-30-second' which hints at the default hookWindowSec behavior, but fails to explain videoIdsOrUrls format expectations or the critical dryRun parameter (which likely controls persistence). Insufficient compensation for missing schema documentation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  States specific action ('Heuristically score') and target resource ('first-30-second hooks across one or more videos'). The specificity of 'hooks' distinguishes it from general analysis siblings like analyzeVideoSet, though it doesn't explicitly contrast with them.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides no guidance on when to use this tool versus alternatives like analyzeVideoSet or inspectVideo. The bracketed timing '[~3-10s, ~1s per video]' suggests performance characteristics but doesn't indicate appropriate use cases or prerequisites.

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

• listChannelCatalogC
  2.8/5.0
  Behavior3/5

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

  No annotations provided, so description carries full burden. Adds valuable timing context [~3-10s] and hints at output format ('compact creator-analysis shape'). However, fails to clarify the 'dryRun' parameter's purpose (suggesting possible side effects not explained) or cache behavior.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness3/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise (single sentence + timing), with no wasted words. However, for a 7-parameter tool with zero schema documentation and no annotations, this brevity becomes underspecification rather than efficiency.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Inadequate for the tool's complexity. With rich filtering options (shorts/long-form, date ranges, sorting) and no output schema, the description omits parameter semantics, error behaviors, and pagination details needed for correct invocation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 0% description coverage with 7 parameters. The description fails to compensate by explaining critical parameters like 'dryRun' (unclear why a list operation needs this), 'publishedWithinDays' bounds, or the enum values for 'sortBy'. Names are intuitive but insufficient for proper usage.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb ('List') and resource ('channel's recent catalog'). The phrase 'compact creator-analysis shape' effectively distinguishes this from siblings like 'inspectChannel' (detailed inspection) or 'buildVideoDossier' (comprehensive reports), though it doesn't explicitly name alternatives.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No guidance on when to use this versus 'inspectChannel', 'findVideos', or 'analyzeVideoSet'. The timing bracket [~3-10s] provides latency expectations but not selection criteria.

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

• researchTagsAndTitlesC
  2.8/5.0
  Behavior2/5

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

  Provides a latency estimate [~3-10s] which is helpful, but with no annotations and no output schema, the description fails to disclose whether the tool performs read-only queries or writes data (despite having a dryRun parameter), what it returns, or any rate limiting concerns.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise with zero fluff; the single sentence and latency annotation are front-loaded and efficient, though the brevity contributes to incompleteness given the lack of structured documentation elsewhere.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Inadequate for a 5-parameter tool with no output schema and no annotations. The description omits what data structure is returned, how maxExamples affects results, and what behavioral constraints (if any) exist beyond the timing hint.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage, the description must compensate but only implicitly references 'seed topic' (mapping to seedTopic). It provides no semantics for regionCode, language, maxExamples, or the critical dryRun flag which suggests potential side effects.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool researches 'title structures, keywords, and tag patterns' using a seed topic, distinguishing it from sibling analysis tools like analyzePlaylist or inspectVideo which focus on existing content rather than pattern research around a topic.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No guidance provided on when to use this versus similar research-oriented siblings like discoverNicheTrends or exploreNicheCompetitors, nor any mention of prerequisites or when not to use it.

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

• analyzePlaylistC
  2.9/5.0
  Behavior3/5

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

  The description valuably discloses performance characteristics '[~5-20s, scales with playlist size]' and the 'partial success' handling model for batch operations. However, lacking annotations, it fails to clarify whether this creates persistent collections, requires specific permissions, or is idempotent.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The single-sentence structure is front-loaded and free of fluff. The bracketed timing notation is information-dense. However, extreme brevity is inappropriate given the tool's complexity and complete absence of structured documentation.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a 6-parameter analytical tool with zero schema descriptions, no annotations, and no output schema, the 20-word description is inadequate. It omits what 'partial success' returns (error objects vs. skipped items), what aggregate benchmarks contain, and how the enum values in 'analyses' differ.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage across 6 parameters, the description fails to compensate. While 'aggregate benchmarks' loosely implies the 'analyses' parameter's purpose, critical parameters like 'dryRun', 'transcriptMode', and 'commentsSampleSize' remain undocumented in both schema and description.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool 'Expand[s] and analyze[s] a playlist' with 'aggregate benchmarks', specifying the resource and actions. However, it lacks explicit differentiation from siblings like 'expandPlaylist' (which likely only expands) or 'analyzeVideoSet' (which analyzes individual videos rather than playlists).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No guidance provided on when to use this composite tool versus chaining 'expandPlaylist' and 'analyzeVideoSet' separately, or when partial success is acceptable versus when atomicity is required.

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

• measureAudienceSentimentC
  2.9/5.0
  Behavior3/5

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

  Discloses critical traits: 'heuristic' nature (non-deterministic), approximate runtime, and specific output components. However, with no annotations provided, it omits safety context like whether it consumes API quota, caches results, or fails gracefully.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely compact single sentence with bracketed latency annotation. No redundant words. However, the brevity is excessive given the complete absence of schema documentation—additional parameter guidance is needed here.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Moderately complex tool (NLP analysis) with no output schema and no annotations. While it lists output categories, it lacks error handling details, data scope (all comments vs. sampled?), or prerequisites (e.g., imported comments required). Insufficient for safe invocation without trial-and-error.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage, the description partially compensates by implying 'themes' and 'quote samples' map to boolean flags. However, it fails to explain 'videoIdOrUrl' format expectations, 'sampleSize' defaults/behavior, or 'dryRun' purpose, leaving 3/5 parameters semantically undefined.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  States specific function (heuristic sentiment analysis from comments) and key outputs (themes, risk signals, quotes). However, it doesn't explicitly differentiate from sibling tools like 'readComments' (raw retrieval) or 'analyzeVideoSet' (broader analysis), leaving selection ambiguous.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides only a latency hint ([~3-10s]) but lacks explicit when-to-use guidance or alternatives. Doesn't clarify when to use this vs. 'searchComments' for filtering or 'readComments' for raw data access.

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

• readCommentsC
  2.9/5.0
  Behavior3/5

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

  Provides useful latency estimate [~1-3s] not found in annotations. However, with no annotations provided, the description omits critical safety context: it doesn't confirm this is read-only (distinguishing it from 'importComments'), doesn't explain what 'structured provenance' contains, and omits error behavior or rate limits.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise single sentence front-loaded with the core action. The timing bracket is efficiently placed. Minor deduction for 'structured provenance' being jargon without explanation.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Inadequate for a 7-parameter tool with zero schema documentation and no output schema. Missing: return value structure, explanation of 'provenance', parameter details, and error conditions. The description covers barely the minimum conceptual surface area.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage, the description partially compensates by implying functionality for 'includeReplies' and 'maxTopLevel' via 'optional replies' and 'top-level'. However, it fails to address 'dryRun', 'order', 'languageHint', or the required 'videoIdOrUrl' parameter semantics, leaving most parameters undocumented.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  States clear verb (read) and resource (comments), specifying 'top-level' and 'optional replies' as scope modifiers. However, it fails to distinguish from sibling 'searchComments'—both read comments, but this appears to be direct retrieval by video ID vs. search across collections.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No guidance on when to use this versus 'searchComments', 'importComments', or other comment-related siblings. The timing notation [~1-3s] hints at latency but doesn't explain selection criteria.

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

• searchTranscriptsC
  2.9/5.0
  Behavior3/5

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

  With no annotations provided, the description must carry the full burden and adds some behavioral context including the '[~instant]' performance marker and default active-collection behavior. However, it fails to confirm read-only safety status, error conditions, or ranking methodology.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description consists of a single efficient sentence that is front-loaded with the core action. The '[~instant]' performance marker is somewhat cryptic but the structure avoids unnecessary verbosity.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given six undocumented parameters and no output schema, the description is inadequate as it leaves most parameters unexplained and provides only minimal detail about the return structure ('ranked timestamped chunks') without clarifying the search syntax or result format.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The input schema has 0% description coverage for its 6 parameters, yet the description only implicitly addresses the collection scoping logic (collectionId/useActiveCollection) without explaining query syntax, the purpose of minScore/videoIdFilter, or maxResults constraints.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description provides a specific verb ('Search') and resource ('imported transcript-text collections'), and clarifies the return format ('ranked timestamped chunks'). It distinguishes scope with 'active-collection focus by default,' though it does not explicitly differentiate from sibling `readTranscript`.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description mentions 'active-collection focus by default,' implying a default scoping behavior, but provides no explicit guidance on when to use this tool versus `readTranscript` or `searchComments`. It lacks prerequisites such as requiring pre-imported collections.

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

• importVideosB
  3/5.0
  Behavior3/5

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

  Includes valuable timing estimate '[~5-20s, ~2s per video]' not found in annotations. Mentions 'local' storage implying disk side effects. However, lacks disclosure on idempotency, error handling (partial failures), or what happens when reindexing existing videos despite the 'reindexExisting' parameter.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely efficient structure: single declarative sentence followed by concise timing metadata in brackets. No filler words or redundant phrases. Information density is high relative to length.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Inadequate for a complex 13-parameter configuration tool. Missing explanation of chunking strategies, embedding provider differences, collection lifecycle (create vs update), and dry-run behavior. Without output schema or annotations, the description should elaborate on these operational dimensions.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage across 13 parameters, the description fails to compensate for the undocumented schema. Only implies the required 'videoIdsOrUrls' parameter ('one or more videos'). Critical configuration options (chunkStrategy, embeddingProvider, dryRun, activateCollection) receive no explanation despite being complex enums and booleans.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  States specific action (import videos), destination (local transcript collection), and purpose (semantic search). Distinguishes from sibling tools like 'readTranscript' or 'importComments' by emphasizing the persistent collection aspect, though could explicitly contrast with transient transcript retrieval.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides no guidance on when to use versus alternatives like 'importPlaylist' (for bulk playlist imports) or 'readTranscript' (for one-off transcript access). No mention of prerequisites like existing collections or video accessibility requirements.

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

• inspectVideoB
  3/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full burden. It successfully discloses the compact nature of returned data and the ~1-3s latency. However, it fails to clarify the safety profile (the presence of a 'dryRun' parameter suggests potential side effects or costs, but this is unexplained) or pagination/error behaviors.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is extremely concise with zero waste—one sentence delivering the core value proposition and a bracketed timing annotation. Information is appropriately front-loaded.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given 4 parameters with 0% schema coverage, no annotations, and no output schema, the description is insufficient. It hints at the purpose of 2 boolean flags but leaves the identifier format and dryRun behavior undocumented, and provides no indication of the return structure despite the lack of output schema.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, requiring the description to fully compensate. While 'normalized ratios' and 'transcript availability' implicitly map to includeEngagementRatios and includeTranscriptMeta, the description omits the videoIdOrUrl format expectations and provides no guidance on the dryRun parameter's function.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description provides a specific verb ('Inspect'), resource ('video'), and clear scope ('single video', 'compact metadata'), which distinguishes it from bulk siblings like analyzeVideoSet and inspectChannel. However, it does not explicitly contrast with these siblings or explain the inspect-vs-analyze distinction.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The '[~1-3s]' timing hint provides implicit guidance about latency expectations, but there is no explicit when-to-use guidance, no mention of prerequisites (e.g., video ID format), and no comparison to alternatives like readTranscript or buildVideoDossier.

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

• listCollectionsB
  3/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full burden. It adds valuable behavioral context with '[~instant]' (performance) and 'local' (scope limitation), but omits mutation status, authentication requirements, error behaviors, or what the return structure looks like.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is appropriately compact with one dense sentence front-loaded with the core action and resources. The '[~instant]' performance tag is concise shorthand, though slightly informal. No wasted words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given this is a simple listing tool with no output schema and only one optional parameter, the description adequately covers the main entities being queried. However, it falls short by not describing the parameter's function or the return value structure, which would be expected for full completeness.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The schema has 0% description coverage for the single boolean parameter 'includeVideoList'. The description fails to mention this parameter or explain what toggling it controls (presumably whether to include full video details versus just counts), leaving its semantics inferred only from the property name.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses the specific verb 'List' with the resource 'local transcript collections' and clarifies what data is returned (active search focus, indexed video/chunk counts). It distinguishes from siblings like 'listCommentCollections' and 'listMediaAssets' by specifying 'transcript' collections, though it could more explicitly contrast with these similar listing tools.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description includes the performance tag '[~instant]' which hints at speed characteristics, but provides no explicit guidance on when to use this tool versus alternatives like 'listCommentCollections' or 'listChannelCatalog', nor any prerequisites or conditions.

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

• readTranscriptB
  3/5.0
  Behavior3/5

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

  No annotations provided, so description carries full burden. Adds valuable latency hint [~1-3s] and mentions 'long-video safeguards' implying chunked processing. However, missing: return format structure, authentication requirements, rate limits, and error handling for videos without transcripts.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely dense single sentence packs mode options, behavioral safeguards, and latency estimate. Every word earns its place. Front-loaded with primary action. No redundant or filler text.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness2/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  With 8 parameters, 0% schema coverage, no output schema, and no annotations, the tool complexity exceeds the description's coverage. Critical gaps remain in parameter semantics and return structure for a multi-modal read operation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0%, requiring heavy description compensation. Mentions modes (mapping to 'mode' enum) and implies pagination/chunking, but leaves 5+ parameters undocumented: dryRun purpose, language code format, includeTimestamps effect, and videoIdOrUrl format expectations.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Specific verb 'Read' + resource 'transcript' with clear mode differentiation (summary, key moments, chapters, full). Distinguishes from sibling searchTranscripts by emphasizing consumption modes vs. search. Could be 5 if it explicitly contrasted with searchTranscripts or inspectVideo.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Lists available modes but provides no guidance on when to use each (e.g., when to choose summary vs. key_moments). Fails to mention when to use vs. sibling searchTranscripts or inspectVideo. No prerequisites or error conditions mentioned.

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

• removeCommentCollectionB
  3/5.0
  Behavior3/5

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

  Adds important context that the operation is 'local' and deletes both the collection and its 'search index', indicating cascading effects. '[~instant]' suggests synchronous execution. However, lacks critical safety disclosure about irreversibility given the destructive nature and absence of annotations.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise with core action front-loaded. '[~instant]' tag is slightly cryptic but efficiently conveys performance characteristic. No redundant text.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for a single-parameter deletion tool, identifying scope (local collection + index). However, given 0% schema coverage and no annotations, it lacks necessary safety warnings about data destruction that would typically be required for irreversible operations.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 0% description coverage. The description fails to compensate by explaining the collectionId parameter, though the parameter name is intuitive given the tool name.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb 'Delete' and specific resource 'local comment collection and its search index'. Distinguishes from generic removeCollection by specifying 'comment' and 'local', though could better differentiate from clearActiveCommentCollection.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this versus clearActiveCommentCollection (which clears the active collection) or removeCollection. No prerequisites or warnings about data loss.

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

• analyzeVideoSetB
  3.1/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full disclosure burden. It successfully communicates critical behavioral traits: partial success handling, item-level error reporting, provenance tracking, and performance characteristics (~5-20s scaling). However, it omits mutation semantics (whether results are cached/persisted), authorization requirements, or side effects.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Every sentence earns its place. The first sentence densely packs purpose, scope, and behavioral traits (partial success/errors/provenance). The bracketed timing annotation [~5-20s, scales with video count] is efficiently appended. No redundancy or filler text.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a tool with moderate complexity (5 parameters, batch operations, multiple analysis types) and zero schema descriptions, the description provides adequate but incomplete coverage. It establishes the operation mode and performance but leaves parameter semantics and return value structure (no output schema exists) undocumented.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, requiring the description to compensate significantly. While 'multiple analyses' loosely hints at the analyses parameter enum, the description provides no guidance on the other four parameters (videoIdsOrUrls constraints, dryRun purpose, transcriptMode options, commentsSampleSize limits) or valid analysis type values.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses specific verb 'Run' and resource 'analyses across a video set', clearly indicating batch processing of multiple videos. The mention of 'partial success, item-level errors' adds distinctive behavioral context. However, it does not explicitly differentiate from siblings like analyzePlaylist (playlist-based vs. explicit video ID list) or inspectVideo (single vs. batch).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this tool versus alternatives. Given the numerous siblings (analyzePlaylist, inspectVideo, readTranscript, etc.), the description fails to specify selection criteria such as 'use this when you have specific video IDs rather than a playlist' or 'use this when you need multiple analysis types in one call'.

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

• inspectChannelB
  3.1/5.0
  Behavior3/5

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

  Adds useful behavioral context with the '[~3-10s]' latency warning not found in annotations. However, fails to explain the 'dryRun' parameter's purpose (unusual for an inspection tool) or disclose whether results are cached, quota costs, or side effects.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely efficient single-sentence description with bracketed performance metadata. Every element earns its place; no redundancy or verbose filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for a simple two-parameter tool: it establishes the core function and performance characteristics. However, significant gaps remain regarding the 'dryRun' parameter's semantics and the structure/contents of the returned analysis (no output schema exists to compensate).

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0%, requiring the description to fully document parameters. While 'channelIdOrHandleOrUrl' is somewhat inferable from the tool name, 'dryRun' is completely undocumented with no explanation of its effect on the operation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb ('Inspect') and resource ('channel') with specific output details ('summary stats and posting cadence heuristics'). However, it doesn't explicitly differentiate from siblings like 'listChannelCatalog' (list vs analyze) or 'inspectVideo' (entity level).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides implicit usage context through latency notation '[~3-10s]', but lacks explicit guidance on when to use this versus 'analyzePlaylist' or 'inspectVideo', and no mention of prerequisites or common workflows.

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

• removeMediaAssetB
  3.2/5.0
  Behavior2/5

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

  Annotations are absent, so description carries full burden. '[~instant]' hints at timing, but missing critical behavioral details: return value (no output schema), irreversibility warning for a destructive operation, and explanation of deleteFiles parameter's disk impact.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Three concise statements with no redundancy. Front-loaded with purpose. '[~instant]' tag is slightly cryptic but serves a behavioral purpose; otherwise efficient structure.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for basic invocation but gaps remain: missing return value description (critical given no output schema), no safety warning for destructive operation, and no explanation of parameter validation rules despite 0 required fields in schema.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 100%, establishing baseline 3. Description adds value by clarifying the 'one or the other' relationship between assetId and videoIdOrUrl, but completely omits deleteFiles parameter semantics.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  States clear verb ('Remove') and resource ('stored media assets'), distinguishing from siblings like removeCollection and removeCommentCollection. However, lacks explicit clarification of what constitutes a 'media asset' versus other removable resources on this server.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides parameter-level guidance ('Specify assetId... or videoIdOrUrl') implying mutual exclusivity, but fails to specify that one parameter is required (schema shows 0 required). No tool-level comparison with alternatives like removeCollection or prerequisites mentioned.

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

• buildVideoDossierB
  3.3/5.0
  Behavior3/5

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

  Adds valuable behavioral context not in annotations: '[~3-10s]' indicates latency cost, and 'one-shot' suggests atomic execution. However, missing critical safety disclosure since annotations are absent—doesn't clarify if 'Build' creates persistent storage, requires auth, or is idempotent.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely efficient single-sentence structure with high information density. The timing annotation '[~3-10s]' is appended without clutter. No filler words or tautology.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for basic comprehension given the tool's complexity (6 params, rich functionality), but gaps remain: no output schema means the description should ideally sketch the dossier structure/return format, and 0% schema coverage leaves half the parameters unexplained.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema description coverage, the description partially compensates by mapping boolean flags to conceptual groups ('core metadata/transcript readiness' vs 'extended with comments, sentiment'). However, it fails to explain 'commentSampleSize', 'dryRun', or input format expectations for 'videoIdOrUrl'.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb ('Build') and resource ('video dossier') with specific scope ('one-shot'). However, it doesn't explicitly differentiate from siblings like 'inspectVideo' or 'analyzeVideoSet' which appear to overlap in functionality.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Implies usage patterns through 'optionally extended with comments, sentiment, and provenance' suggesting when to toggle flags, but provides no explicit guidance on when to choose this tool over alternatives like 'inspectVideo' or when to avoid it.

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

• findVideosB
  3.3/5.0
  Behavior4/5

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

  Excellent given zero annotations: explicitly describes return characteristics ('compact ranked results with provenance and engagement hints') and latency ('[~1-3s]'). Adds meaningful context about result format and performance expectations beyond what any annotation would provide.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely tight: two sentences front-loaded with purpose, specific behavioral traits, and latency hint. Zero wasted words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for basic invocation but insufficient for a 9-parameter tool with complex filtering (date ranges, duration buckets). Missing output schema means description should elaborate on 'provenance' and 'engagement hints' structure, and parameter guidance is needed given low schema coverage.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With only 11% schema coverage (just 'query' described), the description fails to compensate for 8 undocumented parameters including 'dryRun', 'duration' enums, date filters, and 'order' options. 'By intent' vaguely implies query semantics but doesn't clarify filtering capabilities.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb ('Search') and resource ('YouTube videos'), with specific phrasing 'by intent' suggesting semantic search capabilities. Distinguishes sufficiently from siblings like searchComments or searchTranscripts which target specific data types, not whole videos.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit when-to-use guidance or differentiation from siblings like 'inspectVideo' (direct lookup) vs this search functionality. No mention of when to use filtering params vs simple query.

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

• listMediaAssetsB
  3.3/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full burden of behavioral disclosure. It adds the '[~instant]' performance characteristic and describes return values (file paths, sizes, manifest metadata), which is helpful. However, it lacks explicit safety disclosure (read-only status), error behavior, or pagination details despite having a 'limit' parameter.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise with zero wasted words. Front-loaded with the action verb. Four efficient fragments cover purpose, filtering, output fields, and performance. Every sentence earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a 3-parameter list tool with no output schema, the description adequately compensates by describing return values (file paths, sizes, metadata). However, gaps remain: the undocumented 'limit' parameter lacks explanation, and without annotations, the description should explicitly state this is a read-only/safe operation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 67% (limit parameter lacks description). Baseline score is 3. The description mentions 'Filter by video or kind,' which aligns with the two described parameters but adds no syntax details, format specifications, or explanation of the undocumented 'limit' parameter.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses a specific verb ('List') with clear resource ('locally stored media assets') and scope. It distinguishes from sibling 'downloadAsset' via the 'locally stored' qualifier and from 'removeMediaAsset' via the action verb. However, it does not explicitly differentiate from search/index siblings like 'searchVisualContent'.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this tool versus alternatives like 'downloadAsset' (fetch remote vs list local) or 'mediaStoreHealth'. The '[~instant]' tag implies performance characteristics but does not constitute usage guidelines or prerequisites.

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

• recommendUploadWindowsB
  3.3/5.0
  Behavior3/5

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

  No annotations provided, so description carries full burden. The '[~3-10s]' timing disclosure adds valuable latency context not inferable from schema. However, fails to explain the presence of 'dryRun' parameter—leaving ambiguous whether this tool has side effects, creates persistent recommendations, or is purely read-only.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise two-part structure: functional description followed by latency annotation. Every element earns its place; no redundancy. Front-loaded with action and resource.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Lacks output schema, yet description provides no hint about return format (time ranges? specific timestamps? scores?). With no annotations and mysterious 'dryRun' parameter, the description should clarify state changes or persistence. Core function is covered but operational context is incomplete.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is low (25%), requiring description compensation. Mentions 'timezone' explicitly and 'recent publishing history' implicitly maps to lookbackDays, adding semantic context. However, fails to explain 'channelIdOrHandleOrUrl' format expectations or the behavioral implications of 'dryRun', leaving half the parameters undocumented.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear specific verb ('recommend') + resource ('upload windows') + data source ('recent publishing history')._implicitly distinguishes from siblings like analyzePlaylist or inspectChannel by focusing narrowly on upload timing optimization, though lacks explicit contrast.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this versus other analysis tools (e.g., inspectChannel or analyzePlaylist). The '[~3-10s]' indicates performance latency but not functional prerequisites or when to avoid.

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

• removeCollectionB
  3.3/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full disclosure burden. It successfully indicates cascade behavior ('and its search index') and performance characteristics ('[~instant]'). However, it lacks information on reversibility, required permissions, or error conditions.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely concise at two phrases. The core action is front-loaded. The '[~instant]' notation is compact if slightly cryptic. Only deduction is that brevity comes at the cost of parameter documentation.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a single-parameter deletion tool without output schema, the description covers the primary effect adequately. However, the complete lack of parameter documentation (given 0% schema coverage) leaves a significant gap for agent operation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0% and the description fails to compensate by explaining the 'collectionId' parameter—what format it takes, where to obtain it (likely from 'listCollections'), or any constraints. The description mentions 'collection' generally but not the specific parameter.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses a specific verb ('Delete') and clearly identifies the resource ('local transcript collection and its search index'). It effectively distinguishes from siblings like 'removeCommentCollection' (comments vs transcripts) and 'clearActiveCollection' (state clearing vs deletion).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  No explicit guidance on when to use this tool versus alternatives like 'clearActiveCollection' or 'removeCommentCollection'. No prerequisites mentioned (e.g., whether collection must be inactive first) or warnings about data loss.

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

• mediaStoreHealthB
  3.4/5.0
  Behavior3/5

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

  With no annotations, the description carries the full burden. It discloses specific subsystems checked (disk, assets, external binaries) and performance characteristics (~instant), but omits safety profile, return format, and side effects.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single sentence front-loaded with action and resource. '[~instant]' tag is efficient metadata. No redundant text; every element earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Lists inspection targets adequately for a simple health tool, but given lack of output schema, it fails to describe return structure or format. Missing explicit differentiation from 'checkSystemHealth' given the sibling relationship.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Zero parameters present. Per rubric, empty schema with 100% coverage (vacuously) warrants baseline 4. Description appropriately requires no parameter explanation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear verb 'Check' and specific resource 'local media store' with enumerated aspects (disk usage, asset counts, ffmpeg/yt-dlp availability). However, it does not explicitly distinguish from sibling 'checkSystemHealth', leaving selection ambiguity despite different naming.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines2/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides only a performance hint '[~instant]' suggesting fast execution. No explicit guidance on when to use versus 'checkSystemHealth', prerequisites, or operational contexts.

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

• setActiveCollectionA
  3.5/5.0
  Behavior3/5

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

  Adds '[~instant]' timing hint and clarifies this affects transcript search specifically, but omits crucial state-mutation details like persistence duration, error behavior for invalid IDs, and relationship to clearActiveCollection given zero annotation coverage.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single front-loaded sentence with minimal verbosity; '[~instant]' is compact though slightly cryptic, and the sentence efficiently conveys both action and primary scope constraint.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for a simple state-setting tool with one parameter and no output schema, but insufficient given the 0% schema coverage—missing parameter documentation and operational details like state persistence or conflict resolution.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Fails to compensate for 0% schema description coverage; mentions 'collectionId' only in the context of search omission rather than documenting the required parameter's format, valid values, or source (e.g., from listCollections).

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description uses specific verb 'Set' with clear resource 'default collection' and explicitly scopes the effect to 'transcript search', effectively distinguishing it from sibling tool setActiveCommentCollection.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Implies usage context by mentioning 'when collectionId is omitted' in search operations, but lacks explicit guidance on when to use versus clearActiveCollection or prerequisites like obtaining valid IDs from listCollections.

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

• checkSystemHealthA
  3.6/5.0
  Behavior3/5

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

  With no annotations, the description carries the full burden. It adds valuable latency context (~3-10s) and scope clarity, but fails to clarify the mutation semantics of 'dryRun' (whether standard operation has side effects) or what constitutes failure versus success.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single sentence delivers purpose, scope, and timing estimate with zero redundancy. Information is front-loaded and dense—every clause earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for a health check tool given the lack of output schema (return value can be inferred), but incomplete due to undocumented parameters. For a tool with 0% schema coverage, the description should explain the boolean flags to reach full completeness.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 0% description coverage for the two boolean parameters (runLiveChecks, dryRun), and the description provides no compensation. The agent cannot discern whether dryRun prevents writes or if runLiveChecks enables network calls versus cached validation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description provides specific verb ('Check'), clear resource ('setup and provider health'), and enumerates exact components verified (yt-dlp, YouTube API, Gemini embeddings, local storage). This distinguishes it from sibling tools like mediaStoreHealth (narrower scope) and checkImportReadiness (different purpose).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The timing annotation '[~3-10s]' implies when to use (when you can tolerate brief latency), and listing the providers implies use when verifying system connectivity. However, it lacks explicit when-not guidance or comparison to alternatives like mediaStoreHealth.

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

• checkImportReadinessA
  3.8/5.0
  Behavior4/5

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

  Strong behavioral disclosure given zero annotations: specifies exact diagnostic areas (transcript availability, sparse-transcript warnings, yt-dlp/API issues) and provides latency estimate [~1-3s]. Missing only side-effect disclosure (caching, quotas) and read-only vs. destructive classification.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Perfectly efficient single-sentence structure. Front-loaded purpose ('Diagnose whether a video is importable') followed by specific scope details and timing metadata. Every clause earns its place with zero redundancy.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Adequate for a diagnostic tool: covers inspection criteria and timing. However, incomplete due to missing parameter documentation (critical given 0% schema coverage) and no indication of return structure despite complex failure modes (yt-dlp errors, sparse transcripts).

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 0% schema coverage, the description fails to document 2 of 3 parameters (language, dryRun). Only implicitly references videoIdOrUrl via 'whether a video is importable'. Language parameter ambiguity (video language vs. transcript language?) is unresolved.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Excellent specificity with 'Diagnose' verb, 'video' resource, and clear scope ('importable'). Distinguishes effectively from siblings like importVideos (which performs the actual import) and inspectVideo (general metadata) by specifying import-blocker diagnostics (yt-dlp/API issues, sparse transcripts).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides implied usage context by detailing what it diagnoses (transcript availability, API issues), suggesting use before importing. However, lacks explicit guidance on when to use versus inspectVideo or analyzeVideoSet, and doesn't state prerequisites or when-not-to-use conditions.

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

• searchCommentsA
  3.8/5.0
  Behavior4/5

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

  No annotations provided, so description carries full disclosure burden. It effectively compensates by describing return structure (author, like count, relevance score) and performance characteristic ([~instant]). Mentions 'active collection' default state. Missing explicit safety declaration (read-only), though implied by 'search'.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Four tightly constructed sentences. Front-loaded with purpose, followed by return values, default behavior, and performance tag. Zero waste—every clause adds information not present in structured schema fields.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Good coverage given no output schema: explicitly documents return fields. 6 parameters with complex filtering capabilities (videoIdFilter, minScore) but description only covers subset. Appropriate for search tool complexity, though mentioning the video filtering capability would improve completeness.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is low at 33% (only query and collectionId have descriptions). Description compensates partially by explaining 'active collection' concept (covering collectionId/useActiveCollection) and 'ranked results' (hinting at minScore). However, fails to explain videoIdFilter purpose or maxResults default behavior/limits.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear specific verb (Search) + resource (imported comment collections) + behavior (ranked results). Mentions 'imported' implying prerequisite. However, does not explicitly distinguish from sibling 'readComments' (which likely retrieves by ID rather than search ranking).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides valuable default behavior context ('Uses active comment collection by default'). Implicitly suggests prerequisite via 'imported.' However, lacks explicit guidance on when to use vs 'readComments' or that 'importComments' must be run first.

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

• setActiveCommentCollectionA
  3.8/5.0
  Behavior3/5

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

  No annotations provided, so description carries full burden. Adds '[~instant]' performance hint and discloses stateful side effect (modifies searchComments behavior). Omits persistence scope (session vs permanent), validation behavior, and error handling details.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single efficient sentence front-loaded with the action. '[~instant]' tag adds behavioral metadata without verbosity. Zero redundancy.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Appropriate for a simple one-parameter state configuration tool. Core purpose and sibling relationship are clear. Parameter documentation gap and lack of output/error specification prevent higher score given 0% schema coverage.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 0% with one undocumented parameter. Description mentions 'comment collection' generally but fails to describe the collectionId parameter format, valid values, or constraints (e.g., must be from listCommentCollections). Insufficient compensation for schema gaps.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear action 'Set', specific resource 'default comment collection', and explicit scope 'when collectionId is omitted'. Distinguishes from sibling setActiveCollection by specifying 'comment' and from clearActiveCommentCollection by describing the positive configuration action.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Explicitly identifies sibling tool searchComments as the consumer of this setting and describes the specific condition triggering its use (omitted collectionId). Lacks explicit 'when not to use' guidance or prerequisite warnings.

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

• exploreNicheCompetitorsA
  3.9/5.0
  Behavior4/5

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

  Strong disclosure given zero annotations: includes latency indicator [~5-20s], explains the analytical method (search result analysis), and documents the conceptual return payload (channel-level stats, top videos, landscape summary). Missing only auth/rate limit details.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Perfectly efficient: three sentences front-loaded with the core action, followed by return value description and usage context. The [~5-20s] timing notation is appended without clutter. Zero waste.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Appropriately complete for moderate complexity (4 flat parameters, no output schema). Compensates for missing output schema by describing returns conceptually. Only gap is failure to document the two schema-undocumented parameters (maxChannels, dryRun).

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters2/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With only 50% schema coverage (niche and regionCode described; maxChannels and dryRun undocumented), the description fails to compensate. No mention of dryRun's purpose (test mode?) or maxChannels' function (result limiting), leaving critical behavioral parameters unexplained.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Excellent specificity: 'Discover active channels in a niche by analyzing who ranks in YouTube search results' provides a concrete verb (Discover), resource (channels), and method (search ranking analysis). Clearly distinguishes from sibling tools like discoverNicheTrends (trends vs. competitors) and inspectChannel (single channel vs. landscape).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides contextual timing ('before entering a niche') and use-case ('competitive reconnaissance'), but lacks explicit contrast with alternatives like discoverNicheTrends or inspectChannel, and omits when-NOT-to-use guidance.

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

• findSimilarFramesA
  3.9/5.0
  Behavior4/5

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

  Strong behavioral disclosure given zero annotations: includes performance characteristics '[~30-60s, vision comparison]' and return type 'image-backed matches'. However, does not disclose state changes (what dryRun prevents) or failure modes.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Optimal density: opening clause establishes purpose/technology, second clause covers input/output modalities, bracketed suffix provides performance metadata. Zero redundant words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given 6 parameters (50% schema coverage) and no output schema, description covers core functionality and timing but omits return structure details and the three undocumented parameters' semantics. Minimum viable for a CV tool of this complexity.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 50% (3 undocumented: maxResults, minSimilarity, dryRun). Description adds critical semantic value by clarifying assetId and framePath are mutually exclusive alternatives (XOR), but does not compensate for the three undocumented parameters.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Excellent specificity: 'Find frames' (verb) + 'using Apple Vision image feature prints' (method/resource). Clearly distinguishes from text-based sibling 'searchVisualContent' and extraction tool 'extractKeyframes' by emphasizing visual similarity matching.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides implied usage guidance by describing the two alternative input methods (assetId OR framePath), but lacks explicit when-to-use guidance versus 'searchVisualContent' or prerequisites like whether frames must be indexed first.

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

• indexVisualContentA
  3.9/5.0
  Behavior4/5

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

  With no annotations provided, the description carries significant weight. It successfully discloses duration (~30-120s), computational cost (downloads + OCR + vision), and return format (frame evidence with local image paths). However, it omits persistence details—whether the index is stored permanently, if it can be queried later, or implications of `forceReindex`.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Extremely efficient: two sentences deliver purpose, return value, and timing. The bracketed duration note is high-signal. No redundancy or tautology; every clause earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For an 11-parameter mutation tool with no output schema, the description covers the basics but misses the critical relationship to `searchVisualContent` (likely dependent on this). It should explicitly state that this creates a persistent searchable index or clarify one-time vs. cached behavior.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 73%, establishing a baseline of 3. The description mentions 'optional Gemini frame descriptions' and 'downloads,' which loosely map to parameters, but adds minimal semantic detail beyond the schema's own descriptions. It notably fails to explain the `dryRun` parameter (undocumented in schema).

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses specific technical verbs ('Build') and resources ('visual index') that clearly distinguish this from siblings like `extractKeyframes` (extraction-only) and `searchVisualContent` (query-only). It specifies the exact technologies employed (Apple Vision OCR, feature prints, Gemini), leaving no ambiguity about the tool's scope.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  While the timing bracket [~30-120s] hints at cost, the description lacks explicit guidance on when to use this versus `searchVisualContent` (which presumably requires this index) or `extractKeyframes`. It doesn't state prerequisites (e.g., 'use this before searching') or when to avoid re-indexing.

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

• searchVisualContentA
  3.9/5.0
  Behavior4/5

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

  No annotations provided, so description carries full burden. Discloses key behaviors: returns 'frame/image evidence with timestamps', uses specific AI models (Apple Vision, Gemini), and performance characteristics. Missing safety/auth information and exact nature of auto-indexing side effects.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Dense, information-rich single paragraph. Front-loaded with purpose, followed by technologies, return format, and performance expectations. Bracketed timing notation is efficient. Zero redundant words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Complex tool with 14 parameters and no output schema. Description adequately covers core search functionality and return types, but fails to explain several parameters (imageFormat, width, dryRun, download options) that affect invocation. Given the parameter richness, additional guidance on the download behavior and dryRun would be necessary for complete coverage.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is low (36%), requiring description compensation. Successfully adds meaning to includeGeminiDescriptions and includeGeminiEmbeddings via 'optional Gemini frame descriptions/semantic embeddings'. Also contextualizes intervalSec/maxFrames via timing notes. However, leaves 7+ parameters unexplained including critical dryRun and confusing download-related parameters (autoDownload, downloadFormat) that suggest side effects not mentioned.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Specific verb 'Search' + resource 'visual content of a video or your indexed frame library'. Clearly distinguishes from siblings like searchTranscripts (text) by emphasizing 'actual visual content' and specific technologies (Apple Vision OCR, Gemini).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides implicit guidance via performance timing '[~1-3s if indexed, ~60-120s if auto-indexing]', helping decide when to pre-index. However, lacks explicit guidance on when to use vs. findSimilarFrames or extractKeyframes, and doesn't clarify prerequisites like indexing requirements.

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

• clearActiveCollectionA
  4/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full burden. It adds valuable behavioral context via '[~instant]' (performance characteristic) and explains the search scope expansion effect. However, it omits safety details (whether this deletes data vs. just deselecting) and doesn't clarify if this affects search types beyond transcripts.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is a single efficient sentence with a brief performance annotation. It is front-loaded with the action ('Clear the active collection') and contains no redundant or wasted words.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a simple state-reset tool with no parameters and no output schema, the description is reasonably complete. It explains the functional effect on the primary use case (transcript search). A minor gap is the lack of explicit clarification that this operation doesn't destroy the collection data (distinguishing it from removeCollection), but the naming convention makes this reasonably clear.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  The input schema has zero parameters, establishing a baseline score of 4. The description appropriately doesn't invent parameters, and the empty schema with additionalProperties: false is self-documenting.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the specific action ('Clear the active collection') and its functional consequence ('so transcript search fans back out across all collections'). It effectively distinguishes this from siblings like setActiveCollection (inverse operation) and clearActiveCommentCollection (mentions 'transcript' specifically vs. comments).

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies usage context through the explanation of the fan-out effect, suggesting when you'd want to use this (to broaden search scope). However, it lacks explicit 'when to use' guidance or direct comparison to setActiveCollection as the alternative for filtering.

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

• discoverNicheTrendsA
  4/5.0
  Behavior4/5

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

  With no annotations provided, the description carries full disclosure burden. It succeeds by revealing the data source ('YouTube search data'), execution latency ('[~5-15s]'), and claiming 'honest limitations disclosed', though it omits specific mutation/safety characteristics or rate limit details.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is efficiently front-loaded with purpose, followed by specific return values, data provenance, and timing metadata. Every clause earns its place with zero redundancy—optimal information density for agent consumption.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Given the complexity of outputs (saturation analysis, momentum signals) and lack of output schema, the description adequately details return structure and scope. It appropriately omits low-level parameter schemas but could clarify dryRun semantics or failure modes for complete coverage.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is only 40% (3 of 5 parameters lack descriptions). While the description implies 'recent videos' maps to lookbackDays and 'top-performing' hints at maxResults, it fails to explicitly document dryRun behavior or the interaction between lookbackDays and momentum calculations, providing only partial compensation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description opens with the specific action 'Discover' applied to 'what's trending in a niche right now', and immediately distinguishes itself from sibling tools like 'findVideos' or 'exploreNicheCompetitors' by detailing specific analytical outputs: momentum signals, saturation analysis, content gaps, and format breakdowns.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implies temporal context ('right now', 'momentum signals', 'recent videos') suggesting use for current trend analysis, but provides no explicit when-to-use guidance, alternatives for historical analysis, or cautions against using for saturated markets.

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

• importCommentsA
  4.1/5.0
  Behavior4/5

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

  No annotations provided, so description carries full burden. Provides valuable timing estimate '[~3-10s]', mentions side effects ('into the local comment knowledge base', 'indexes them'), and external dependency ('existing comment pipeline'). Missing idempotency details and error handling behavior.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Two dense sentences plus timing metadata. Every element serves a purpose: first defines action/scope, second explains mechanism/relationship, bracket provides UX-critical latency expectations. Zero redundancy.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness3/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a 9-parameter indexing tool with complex lifecycle (collections, activation, dry-run), the description covers core import mechanics but omits explanation of collection management workflow, dryRun behavior, and return value structure (no output schema present to compensate).

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 67% (>50%), establishing baseline 3. Description adds no explicit parameter guidance to compensate for undocumented fields like 'dryRun' and 'maxRepliesPerThread'. Mentions 'video' implicitly mapping to videoIdOrUrl, but adds no syntax or format details beyond schema.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Specific verb 'Import' with clear resource 'video's comments' and destination 'local comment knowledge base'. Explicitly mentions downstream purpose 'for semantic search' and target sibling 'searchComments', effectively distinguishing it from readComments.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Implies usage context by referencing 'searchComments' as the consumer of the indexed data, suggesting when to use this (when search is needed later). However, lacks explicit 'when not to use' guidance or direct comparison to readComments for immediate viewing without persistence.

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

• clearActiveCommentCollectionA
  4.2/5.0
  Behavior4/5

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

  With no annotations provided, the description carries the full disclosure burden. It successfully explains the behavioral consequence (search scope changes to span all collections) and includes performance characteristics ('[~instant]'), indicating synchronous/fast execution. It implies this is a state reset rather than deletion (distinguishing from removeCommentCollection).

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Single efficient sentence with zero waste. The action is front-loaded ('Clear the active comment collection'), followed by purpose clause ('so...'), and performance tag ('[~instant]'). Every element earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a zero-parameter state management tool without output schema, the description adequately covers the essential context: what it does, the immediate effect on system behavior (search scope), and performance characteristics. It could be improved by explicitly naming affected operations like searchComments or readComments, but 'comment search' provides sufficient context.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Baseline score of 4 applies as there are zero parameters. The description does not need to compensate for parameter documentation since the input schema is trivial (empty object with additionalProperties: false).

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description uses a specific verb ('Clear') and identifies the exact resource ('active comment collection'). It distinguishes from the sibling tool 'clearActiveCollection' by specifying this is for comments, and explains the functional effect on search scope ('fans back out across all comment collections'), clearly differentiating it from 'setActiveCommentCollection'.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines3/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Usage is implied through the explanation of effect ('so comment search fans back out...'), suggesting when to use it (when you want global search scope). However, it lacks explicit guidance like 'Use this when you want to search across all collections instead of a specific one' or naming the alternative 'setActiveCommentCollection' for comparison.

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

• downloadAssetA
  4.4/5.0
  Behavior5/5

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

  Exceptional disclosure given no annotations: specifies return value 'asset manifest entry with file path', performance characteristics '[~30-120s]', side effects 'downloads media', and negative capability 'Does NOT perform visual indexing'. Covers timing, output format, and mutation behavior comprehensively.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Perfectly structured four-sentence description: purpose (sentence 1), return value (2), sibling differentiation (3), performance metadata (4). Zero waste, front-loaded with critical information, appropriate density for complexity.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Comprehensive for a download utility: covers I/O behavior, performance expectations, and sibling distinctions. Missing only minor lifecycle context (relationship to removeMediaAsset for cleanup) and storage persistence policies, but complete for invocation decision.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema has 100% description coverage with detailed enum descriptions ('best_video = highest quality video+audio'). Description provides semantic context mapping to 'video, audio track, or thumbnail' but largely mirrors schema content. Baseline 3 appropriate when schema carries full documentation burden.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Excellent specificity with clear verb 'Download', explicit resources 'YouTube video, audio track, or thumbnail', and destination 'local storage'. Distinguishes from siblings like indexVisualContent and importVideos through explicit scope definition.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Strong differentiation via 'Does NOT perform visual indexing — this is honest file storage', clearly signaling this is distinct from indexVisualContent sibling. However, lacks explicit guidance on when to use versus importVideos or importPlaylist siblings which also handle video ingestion.

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

• extractKeyframesA
  4.4/5.0
  Behavior4/5

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

  No annotations provided, but description carries substantial load: discloses external dependency (requires ffmpeg), execution duration (~30-60s), and output nature (raw frame images). Could specify if operation is idempotent or disk space implications.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Four distinct information units in tight prose: action, prerequisite, negative capability, performance metadata. Every sentence earns its place. Front-loaded with core extraction purpose.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Strong coverage for a processing tool with no output schema: workflow integration (downloadAsset), implementation detail (ffmpeg), timing, and sibling differentiation present. Missing only error condition details or explicit return value description.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 100% with complete parameter descriptions. Description reinforces semantics ('regular intervals' for intervalSec, 'locally downloaded' for videoIdOrUrl constraint) but schema does heavy lifting. Baseline 3 appropriate for high-coverage schemas.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Clear specific verb (Extract) + resource (keyframe images) + method (ffmpeg) + scope (regular intervals). Distinguishes from visual analysis siblings (searchVisualContent, findSimilarFrames) by stating it 'Does NOT do visual search or classification'.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines5/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Explicit prerequisite workflow ('Requires the video to be downloaded first via downloadAsset') naming specific sibling tool. Negative constraint clarifies it's for raw extraction, not analysis. Clear when-to-use vs. alternatives.

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

• exploreYouTubeA
  4.8/5.0
  Behavior5/5

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

  With no annotations provided, the description carries significant behavioral weight: it discloses processing approach ('intelligent multi-query ranking', 'parallel enrichment'), latency costs ('[~2s quick, ~5-10s standard, ~15-30s deep]'), return structure ('ranked videos with transcript summaries, key moments'), and crucial agent-side obligations ('ALWAYS create visual charts... immediately').

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Though lengthy (8+ sentences), the description is well-structured with critical priority instructions front-loaded ('ALWAYS use this tool FIRST'), followed by prohibitions, core functionality, return values, mandatory side effects, and parameter guidance. Every sentence provides actionable guidance for a complex 12-parameter tool.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness5/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  Excellent completeness given no output schema exists. Describes return values ('ranked videos with transcript summaries'), processing characteristics ('parallel enrichment'), timing implications for agent planning, and mandatory follow-up behaviors (chart creation), covering the gaps left by missing structured output definitions.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  With 92% schema coverage, the baseline is high, but the description adds valuable semantic context for enum parameters (explaining 'specific' vs 'explore' modes and the three depth levels with concrete outputs for each) and reinforces the query/searches mutual exclusivity implied in the schema.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the core action ('Searches YouTube directly with intelligent multi-query ranking') and explicitly distinguishes this tool from web search ('Do NOT use web search... use this tool instead') and implies distinction from analysis-focused siblings like analyzeVideoSet by emphasizing it's for discovery/exploration and should be used FIRST.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines5/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  Provides explicit when-to-use ('ALWAYS use this tool FIRST when the user wants to find, discover, or explore YouTube videos'), clear alternative exclusion (web search), and detailed parameter guidance ('Use specific mode to find one best video', 'Use explore for topic discovery') including timing expectations for depth settings.

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