Server Quality Checklist

Tool Scores

• crm_delete_contactA
  4.7/5.0
  Behavior5/5

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

  The description goes far beyond the annotations (readOnlyHint: false, destructiveHint: true, idempotentHint: true, openWorldHint: true) by disclosing the soft-delete nature, required OAuth scope ('crm.objects.contacts.write'), idempotent behavior, cache invalidation, and not-found error handling. This gives the agent a full understanding of side effects and prerequisites.

  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 front-loaded with a clear one-sentence summary, followed by usage guidance, behavioral details, permissions, and error cases. Each sentence adds unique value; there is no fluff or repetition. Despite being a bit long, it's well-organized and efficient for a destructive 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?

  Given the tool's simplicity (one parameter), the description covers all critical aspects: purpose, when to use, safety, permissions, idempotency, cache effects, and error behavior. The output schema exists, so return details are not needed. This is thoroughly complete for an agent to safely invoke the tool.

  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?

  The input schema already fully describes contact_id as 'HubSpot contact record id, as returned by search, list, or create,' achieving 100% schema description coverage. The description adds no new parameter semantics beyond noting that an unknown id raises a not-found error, which is minor.

  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 'Archive (soft-delete) a HubSpot contact so it drops out of active views,' which precisely states the action (archive/soft-delete), resource (HubSpot contact), and effect. It clearly distinguishes from sibling tools like crm_get_contact or crm_update_contact by focusing on deletion behavior.

  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?

  Explicitly says 'Use this instead of asking a human to delete in the UI,' giving a clear when-to-use directive. It also notes the tool 'never issues a hard delete' and that records are restorable for a limited window, which helps the agent decide if this is appropriate for the task.

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

• crm_delete_dealA
  4.7/5.0
  Behavior5/5

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

  Beyond the annotations, the description exposes key behaviors: soft-delete semantics, dropping from the read cache and search results, idempotency (archiving again leaves it archived), and not-found errors for never-existing IDs. It also reinforces the destructive nature from the agent's perspective while aligning with the destructiveHint and idempotentHint annotations.

  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 structured with the core action first, then usage guidance, then behavioral details. Every sentence earns its place; the length is justified for a destructive operation with important caveats.

  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?

  Given the tool's destructive nature, annotations, and full schema coverage, the description covers all essential context: scope requirements, error behavior, idempotency, and cache/search effects. The presence of an output schema means returning the archive confirmation is already handled, so no return-value explanation is needed.

  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?

  The single parameter is fully documented in the schema (100% coverage), so the description doesn't need to add parameter syntax. It does add useful context by explaining repeated calls with the same id are idempotent, but the schema already provides the essential meaning.

  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 tool archives (soft-deletes) a HubSpot deal so it disappears from the pipeline board. It distinguishes itself from the sibling crm_update_deal by explicitly contrasting archiving with moving to a closed-lost stage.

  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?

  It gives explicit guidance: prefer crm_update_deal for moving to closed-lost, and only archive when the record should not have existed. It also notes the required OAuth scope, providing clear when-to-use and when-not-to-use context.

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

• crm_export_audit_logA
  4.7/5.0
  Behavior5/5

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

  The description goes far beyond the annotations. It discloses that the tool is entirely local and reads an in-memory buffer, makes no HubSpot call, requires no scope, includes even denied/error calls, redacts email/phone, and explains the buffer is not persisted across restarts. These details fully characterize behavior, aligning with and expanding upon the read-only and idempotent hints.

  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 well-structured and efficient. The first sentence states the core function, the second provides use cases, and the remaining sentences deliver behavioral details without redundancy. Every sentence earns its place, making it concise despite being detailed.

  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?

  Even with no parameters and an output schema present, the description covers all necessary context: purpose, use cases, input (none), output format and content, behavioral traits, redaction, and session-scope limitations. It is fully self-contained and leaves no significant gaps for the agent.

  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 zero parameters, the baseline is 4. The description doesn't need to explain parameter semantics; it adds value by describing the output format and content, which is appropriate for a no-input tool.

  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 tool exports the session's audit trail as JSON Lines with one record per line. The verb 'export' and resource 'audit trail' are specific, and the format 'JSON Lines' adds clarity. This is distinct from sibling CRM tools that focus on contacts, deals, and pipelines.

  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?

  The description provides explicit when-to-use guidance: 'Use this to hand a reviewer evidence... or to attach the trail to a ticket or change record.' It does not mention exclusions or alternative tools, though no alternatives likely exist in the given sibling set, so the absence is not a significant gap.

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

• crm_get_contactA
  4.7/5.0
  Behavior5/5

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

  Beyond the annotations (readOnly, idempotent, not destructive), the description adds crucial operational context: required OAuth scope, in-process cache with 300s TTL, write-invalidation behavior, audit log implications, and the not-found error for unknown/archived ids. No contradiction with annotations.

  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 front-loaded with the core action, then adds usage guidance and operational details. Every sentence contributes essential information—retrieval, prerequisite usage, permissions, caching, invalidation, and error behavior—with no redundancy or fluff.

  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?

  Given the simple one-parameter schema, available output schema, and strong annotations, the description covers all necessary operational nuances: when to use, required scope, caching behavior, and error handling. The agent has complete guidance without needing to infer undefined 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?

  The schema already fully describes the single parameter (contact_id) with its type and source ('as returned by search, list, or create'). The description's phrase 'by record id' adds no additional meaning beyond the schema, so a baseline 3 is appropriate.

  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 tool fetches one HubSpot contact by record id, specifying the returned fields (email, name, phone, company). It distinguishes itself from sibling tools like search, list, create, and delete by focusing on retrieval of a known id.

  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?

  Explicitly instructs to use the tool only after an id is known and names the alternative (crm_search_contacts) to find the id first. This provides clear sequencing and differentiates from the search tool.

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

• crm_get_dealA
  4.7/5.0
  Behavior5/5

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

  Beyond the annotations (readOnlyHint, idempotentHint, destructiveHint), the description adds important context: required OAuth scope, in-process caching with 300s TTL, cache invalidation on writes, and not-found behavior for unknown/archived ids. These details significantly exceed 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.

  Conciseness5/5

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

  The description is concise and well-structured, using short paragraphs to separate purpose, usage guidance, and operational details. Every sentence adds value, with no redundancy or filler.

  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?

  For a simple one-parameter read tool with a rich output schema, the description covers all essential context: purpose, usage, auth requirements, caching behavior, and error conditions. Nothing important is missing.

  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 only one parameter and 100% schema description coverage, the schema already fully documents deal_id. The description adds no further semantic detail beyond what is in the schema, so the baseline score of 3 is appropriate.

  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 ('Fetch') and resource ('one HubSpot deal by record id'), and explicitly lists the fields returned (name, amount, pipeline, stage). It clearly distinguishes from crm_search_deals by stating this tool is used once an id is known.

  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?

  It explicitly states when to use the tool ('once an id is known') and directs users to use crm_search_deals to find the id first, naming the alternative. This is clear and actionable guidance.

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

• crm_get_pipelineA
  4.7/5.0
  Behavior5/5

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

  Annotations already declare readOnlyHint/idempotentHint, but the description adds valuable context beyond that: requires 'crm.schemas.deals.read' scope, is 'not cached and always reaches HubSpot' (unlike the pipeline list), and 'Raises a not-found error for an unknown id.' These are meaningful behavioral traits not present in annotations. No contradiction.

  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 concise and front-loaded with the core purpose. Each sentence adds value: purpose, usage guidance, discovery alternative, and behavioral caveats. No filler or redundancy. Despite several sentences, all are necessary for correct tool use.

  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?

  For a simple one-parameter tool with an output schema and strong annotations, the description covers purpose, when to use, auth scope, caching/network behavior, and error handling. It is fully complete for the task context; nothing critical is missing.

  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% coverage for the single parameter with description 'Deal pipeline id, as returned by crm_list_pipelines.' The tool description repeats this context ('pipeline id is already known', 'by id') but adds little beyond the schema. With high schema coverage, baseline is 3; description offers no extra semantics.

  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 starts with 'Fetch one deal pipeline by id with its stages sorted by display order,' a specific verb+resource combination that clearly states the tool's function. It also distinguishes from sibling tool crm_list_pipelines by focusing on fetching a single pipeline by ID, not listing all pipelines.

  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 guidance: 'Use this when the pipeline id is already known and only its stages are needed.' Also names the alternative: 'Use crm_list_pipelines to discover ids.' This directly addresses tool selection and differentiates from the list sibling.

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

• crm_list_pipelinesA
  4.7/5.0
  Behavior5/5

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

  Beyond the readOnlyHint annotation, the description adds valuable context: required OAuth scope, in-process cache with 300-second TTL, and empty-list semantics. It also explicitly says 'Read-only' reinforcing the safe nature of the call.

  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 compact yet information-dense. The first sentence states the purpose, the second gives usage guidance, and the third covers behavioral details. Each sentence earns its place with no filler.

  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?

  For a simple zero-parameter tool, the description fully covers what the tool returns, when to use it, auth requirements, caching behavior, and edge cases. The existence of an output schema covers return format details, so nothing critical is missing.

  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 tool takes no parameters, so schema coverage is trivially 100%. The description adds no parameter details because none exist, but the baseline of 4 is appropriate for a zero-parameter tool.

  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 a specific verb and resource: 'List every deal pipeline in the account with its stages in display order.' It clearly distinguishes from sibling tools like crm_get_pipeline by indicating it covers all pipelines and includes stage ordering.

  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?

  The description explicitly states when to use this tool ('Call this before creating or moving a deal') and explains the necessity for pipeline and dealstage ids. It doesn't explicitly mention alternatives (e.g., crm_get_pipeline for a single pipeline), but the use case is clearly defined, making it a strong 4.

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

• crm_search_contactsA
  4.7/5.0
  Behavior5/5

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

  Beyond the readOnly/idempotent annotations, the description discloses the required OAuth scope, denial/audit behavior, lack of caching, pagination cursor semantics, and archived contact exclusion. No contradictions with annotations; the description adds substantial behavioral context.

  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 compact and front-loaded: the first sentence states the core purpose, the second gives usage context, the third covers auth and caching, and the fourth explains pagination. Every sentence earns its place with no redundancy.

  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?

  With rich schema annotations, output schema present, and the description covering auth, safety, pagination, and filtering behavior, the agent has everything needed to select and invoke the tool correctly. No critical gaps identified.

  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?

  The input schema already describes all three parameters fully (100% coverage), including defaults and meanings. The description reinforces free-text search and 'after' cursor usage but does not add new semantics beyond the schema, so a baseline of 3 is appropriate.

  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 a clear verb+resource+scope: 'Search HubSpot contacts by free-text term and return one page of matches.' It explicitly contrasts with sibling tools in the usage guidance, distinguishing it from crm_list_contacts (no term) and crm_get_contact (known ID).

  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 guidance: 'Use this to find contacts by name, email, company, or any other indexed property.' It names alternatives: 'Prefer crm_list_contacts when there is no term to match on, and crm_get_contact when the record id is already known.' Also includes operational guidance on pagination and caching.

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

• crm_update_dealA
  4.7/5.0
  Behavior5/5

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

  Beyond annotations (readOnlyHint=false, destructiveHint=true, idempotentHint=true), the description adds rich behavior: 'Writes' status, partial overwrite semantics, retry safety, required OAuth scope 'crm.objects.deals.write', cache invalidation, and specific error conditions (not-found for unknown/archived id, validation for no property). This fully complements and expands on structured hints without contradicting them.

  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 appropriately dense, starting with a clear one-sentence purpose, followed by usage guidance, then a concise paragraph on behavior. Every sentence adds distinct value—no repetition of schema or annotation data, no filler.

  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?

  With an output schema present, return values need not be described. The description covers auth requirements, cache effects, error scenarios, idempotency, and partial-update behavior, making it fully sufficient for an agent to invoke the tool correctly. The tool's complexity is moderate, and all relevant context is provided.

  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?

  Schema covers 100% of parameters, so the baseline is 3. The description adds meaningful context beyond schemas: dealstage must come from the chosen pipeline and HubSpot rejects mismatched stages, and close_date maps to HubSpot's 'closedate' property. This extra information elevates the score.

  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 'Update properties on an existing HubSpot deal' with a specific verb and resource, and explicitly distinguishes from other operations by noting the pipeline cannot be changed here and to 'create the deal in the right pipeline instead'. This separates it from create, delete, get, and search siblings.

  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?

  Provides explicit guidance: 'Use this to advance a deal through its pipeline' and warns that changing the pipeline itself requires creating a new deal. Also explains partial-update semantics (only passed properties are sent, omitted ones untouched), which helps the agent decide when to use this tool. Does not explicitly list all alternatives but covers the main use case.

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

• crm_batch_create_contactsA
  4.9/5.0
  Behavior5/5

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

  It discloses the required write scope, the response shape with 'succeeded' and 'failed' arrays, the non-idempotent behavior, and cache invalidation — all beyond the annotations. This gives the agent a full behavioral model.

  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 about five sentences, each with a distinct purpose: purpose, usage, behavior, idempotency caveat, cache effect. No fluff.

  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?

  For a batch mutation tool, it covers permissions, partial failure semantics, retry guidance, and response structure, so the agent can invoke it safely. The output schema is not shown but the description adequately describes the return value.

  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 schema already describes the contacts array with 100% coverage, and the description adds that failures include the input index and original row, making the parameter's ordering meaningful. It also refers to 'the same batch' which contextualizes the array as a unit.

  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 'Create many HubSpot contacts in one request and report the outcome of every row,' which is a specific action and resource. It also contrasts with crm_create_contact by emphasizing 'many... in one request,' making the distinction clear.

  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?

  It explicitly states 'Use this for bulk loads' and advises using crm_create_contact when a retry must not risk duplicates, giving clear when-to-use and when-not-to-use guidance. No other tool needs that exclusion.

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

• crm_create_contactA
  4.9/5.0
  Behavior5/5

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

  Beyond the annotations (readOnlyHint=false, idempotentHint=true), the description discloses the required OAuth scope, the session-scoped idempotency store, cache invalidation on writes, and the validation requirement of at least one property. No contradiction with annotations; it meaningfully enriches them.

  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?

  Three tight paragraphs: purpose, usage guidance, and behavioral details. Every sentence carries information, with no filler or repetition of the schema. Front-loaded with the core purpose.

  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?

  With an output schema present, the description appropriately avoids re-explaining return values. It covers authentication, idempotency semantics, cache effects, and validation constraints, making it sufficient for an agent to invoke correctly.

  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?

  Schema description coverage is 100%, so the baseline is 3. The description adds valuable semantic context beyond the schema: the 'at least one property' validation rule and the idempotency key derivation/replay behavior. This justifies a 4.

  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 a specific verb and resource: 'Create one HubSpot contact from the properties given and return the new record.' It explicitly contrasts with sibling tools like crm_batch_create_contacts and crm_update_contact, making the tool's unique role unambiguous.

  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 guidance: 'Use this for single creates, especially ones a retry might repeat.' It also names alternatives: crm_batch_create_contacts for bulk loading and crm_update_contact for existing records, giving clear decision criteria.

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

• crm_create_dealA
  4.9/5.0
  Behavior5/5

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

  The description discloses substantial behavioral traits beyond the annotations: it states the operation is 'Writes, additively,' explains the idempotency key mechanism, the in-process key store limitation ('covers retries within a session and not restarts'), cache invalidation, and the minimum property requirement. These details go well beyond the annotations' idempotentHint and destructiveHint flags.

  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 effectively structured into three paragraphs: purpose, prerequisite guidance, and behavioral details. Every sentence provides necessary information with zero fluff. It is appropriately sized for the complexity of the 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?

  Given the tool's complexity—6 optional parameters, external dependency on crm_list_pipelines, OAuth scope requirements, idempotency semantics, and cache invalidation—the description covers all essential context. Return values are covered by the output schema, so no further explanation is needed.

  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 already provides strong descriptions for all 6 parameters (100% coverage), so the baseline is 3. The description adds critical cross-parameter semantics, noting that HubSpot 'only accepts a stage id that belongs to the chosen pipeline' and explaining idempotency_key behavior ('or by a hash of the properties when none is given'). This adds meaningful meaning beyond 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 tool's function: 'Create one HubSpot deal from the properties given and return the new record.' It uses a specific verb and resource, and distinguishes itself from the sibling crm_update_deal by explicitly saying 'Use crm_update_deal to move an existing deal.'

  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 guidance with a prerequisite call: 'Call crm_list_pipelines first when setting pipeline or dealstage,' and directs users to an alternative tool: 'Use crm_update_deal to move an existing deal.' Also states the required OAuth scope and the condition 'At least one property must be supplied.'

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

• crm_list_contactsA
  4.9/5.0
  Behavior5/5

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

  Beyond the annotations (readOnlyHint, idempotentHint, openWorldHint, destructiveHint false), it adds meaningful behavioral context: required OAuth scope, local denial with audit logging, no caching, archived contacts excluded, and cursor-based pagination termination. This goes well beyond the structured metadata.

  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 compact and front-loaded, with each sentence earning its place. It mixes usage guidance, auth requirements, caching behavior, and pagination instructions without redundancy or filler.

  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?

  Despite having an output schema (so return values are covered), the description covers all operational essentials: iteration pattern, auth, rate-limit trade-off, archived-contact filtering, and pagination. This is complete for a paged list tool of moderate complexity.

  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?

  Schema coverage is 100%, so the baseline is 3. The description adds value by explaining how the 'after' cursor is used across pages and when to stop ('stop when it comes back null'), which complements the schema. It does not add much for 'limit', but the schema already documents its range and default.

  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 a specific verb and resource: 'List HubSpot contacts in record-id order, one page at a time.' It clearly identifies the operation and distinguishes itself from crm_search_contacts by explicitly framing this as the paged walk-all tool.

  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?

  The description says exactly when to use it ('to walk the whole contact set') and when not to: 'Prefer crm_search_contacts when there is a term to match on, since listing everything is slower and burns rate limit.' It names the alternative and gives the reason.

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

• crm_search_dealsA
  4.9/5.0
  Behavior5/5

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

  The description adds significant behavioral context beyond the annotations: required OAuth scope ('crm.objects.deals.read'), failure behavior ('denied locally and recorded in the audit log'), caching behavior ('Not cached'), pagination mechanism ('Pass the returned after cursor back for the next page and stop when it is null'), and filtering ('Archived deals are excluded'). No contradiction with the readOnlyHint annotation.

  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 four sentences, each serving a clear purpose: main capability, usage guidance, permission/pagination/filtering details. There is no fluff; it is front-loaded with the core purpose and tightly packed with actionable information.

  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?

  Given the tool's moderate complexity, the description covers all necessary context: it explains pagination, permission requirements, exclusions, and the distinction from sibling tools. An output schema exists, so not detailing return values is acceptable. This is a complete description for an AI agent to select and invoke the tool correctly.

  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?

  Schema coverage is 100%, so the baseline is 3. The description goes beyond the schema by explaining what 'query' matches ('name, amount, or stage text') and how to use the 'after' cursor for pagination. This extra context justifies a 4, though the schema already documents the parameters well.

  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 first sentence states a specific action: 'Search HubSpot deals by free-text term and return one page of matches,' clearly distinguishing this from siblings by specifying the resource (deals) and method (free-text search). It also explicitly contrasts with crm_get_deal and crm_list_pipelines, making the tool's unique role unambiguous.

  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?

  The description provides explicit guidance: 'Use this to find deals by name, amount, or stage text' and then names alternatives: 'Use crm_get_deal when the record id is already known, and crm_list_pipelines to translate the returned pipeline and stage ids into readable labels.' This clearly answers when to use this tool versus others.

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

• crm_update_contactA
  4.9/5.0
  Behavior5/5

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

  Adds significant detail beyond annotations: states it's a write operation, only passed properties are sent, previous values are lost, omitted properties are untouched, retries are safe, requires the write scope, invalidates cache, and raises specific errors. Consistent with annotations; no contradiction.

  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 compact yet comprehensive, covering purpose, usage, behavior, scope, cache, and errors in a logical order without redundancy. Every sentence adds value.

  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?

  For a mutation tool with 6 parameters and an output schema, the description is complete. It addresses when to use, how it behaves, required permissions, error cases, and side effects. The output schema covers return details, so no need to elaborate on that.

  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?

  Schema coverage is 100%, so descriptions already exist per parameter. The tool description adds general semantics about partial updates and overwrite behavior, which enriches understanding of how all parameters behave.

  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 'Update properties on an existing HubSpot contact and return the updated record,' with a specific verb and resource. It also distinguishes from the sibling create tool, saying 'Use crm_create_contact for records that do not exist yet.'

  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?

  Explicitly says 'Use this to change a known record' and directs to crm_create_contact for new records. Error cases (unknown/archived id, no property supplied) provide additional context on when the tool is appropriate.

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