Server Quality Checklist

Tool Scores

• linkedin_list_pending_invitesA
  4.4/5.0
  Behavior5/5

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

  The description explicitly states the tool never accepts, ignores, or withdraws anything, and takes no `confirm`, adding real behavioral context beyond the readOnlyHint annotation. It also discloses edge cases such as empty lists being normal and the difference between `total` and `returned`.

  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 dense but every sentence earns its place: purpose, parameter semantics, read-only guarantee, output format, and edge-case handling. It is front-loaded with the core action and then efficiently covers nuances without 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 no output schema, the description fully explains what entries contain, how timestamps are represented, that an empty list is valid, and how `total` relates to `returned`. Combined with parameter coverage and side-effect disclosure, nothing critical is missing for calling this tool correctly.

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

  Parameters5/5

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

  Schema description coverage is 0%, so the description carries full responsibility for explaining parameters. It thoroughly documents the `direction` enum values with defaults, `count` default and maximum, and even explains result-field meanings that relate to the parameters, fully compensating for the schema gap.

  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 lists 'invitations that are still awaiting a decision' and specifies the resource precisely. It also distinguishes sent vs received directions, but does not explicitly differentiate from the similar sibling tool `linkedin_list_connections`, so it falls just short of a 5.

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

  Usage Guidelines3/5

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

  The description gives strong operational guidance: direction defaults, count limits, and read-only behavior. However, it never mentions when to prefer this tool over alternatives like `linkedin_list_connections` or `linkedin_send_connection_request`, leaving selection context implied rather than explicit.

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

• linkedin_create_postA
  4.7/5.0
  Behavior5/5

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

  Annotations already mark destructive=true, but the description adds substantial context: the two-step preview/publish mechanism, daily posts cap, visibility default, media refusal with invalid_input, 1300-character collapse warning, and dry-run behavior. This goes well beyond what annotations provide and contains no contradictions.

  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?

  While long, every sentence adds operational value: purpose first, then invocation pattern, then limitations and edge cases. There is no filler or redundant phrasing; the structure logically leads the agent through what to expect.

  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?

  Covers the preview response fields, error handling for media, warning threshold, and dry-run behavior. The only notable gap is that it doesn't describe what the final publish call returns (e.g., post ID or confirmation response), which would be useful since there is no output schema. Still, this is minor relative to the volume of useful context.

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

  Parameters5/5

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

  Schema has zero descriptions, so the description carries the full burden. It explains text (content), confirm (publish gate), mediaUrl (accepted but unsupported, refused on confirm), visibility (default 'public', meaning of 'connections'), and previewToken (returned from preview). All parameters are meaningfully clarified.

  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: 'Publishes a text post to the signed-in member’s LinkedIn feed.' This clearly distinguishes it from siblings like send_message or scrape_profile and leaves no ambiguity about what the tool does.

  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?

  It explains the two-step invocation pattern (preview without confirm, publish with confirm) and the media limitation as a clear 'not supported' case with manual workaround. It doesn't explicitly mention sibling alternatives, but the usage context and when-to-publish guidance are strong.

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

• linkedin_list_connectionsA
  4.7/5.0
  Behavior5/5

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

  Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint, so the bar is lower. The description adds significant behavioral detail beyond that: pagination mechanics (scroll-paginated until count met, stops growing, internal cap), how the query filter is applied locally, the total count source behavior, and the per-entry fields including 'connected at' timestamp. This goes well beyond the structured hints.

  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 dense but efficient—about 5 sentences covering all important aspects. It is front-loaded with the core purpose and sorted order, then details parameters and output. Could be slightly more concise (e.g., merging some clauses) but every sentence adds value without 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 tool with only 2 optional parameters and no output schema, the description covers all necessary context: purpose, sorting, pagination, parameter semantics, output fields, total count source, and read-only nature. An agent can confidently call it and interpret results without needing additional external knowledge.

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

  Parameters5/5

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

  Schema coverage is 0%, so the description must fully explain both parameters. It does: count is given a default (50), maximum (200), and describes pagination; query is defined as a case-insensitive substring filter applied locally to name and headline. This fully compensates for the lack of schema descriptions.

  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 lists the signed-in member's 1st-degree connections with a specific sort order (most recently added first). This distinguishes it from siblings like linkedin_send_connection_request (mutating) and linkedin_scrape_profile (single profile), leaving no ambiguity about what it does.

  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 explains when to use it (listing connections) and clarifies that the query filter is local, not LinkedIn's search, which warns against misuse for large lists. However, it does not explicitly point to alternative tools for actions like sending requests or scraping profiles, though the sibling names make that implicit. Slight gap in explicit exclusions.

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

• linkedin_loginA
  4.7/5.0
  Behavior5/5

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

  Annotations already indicate readOnlyHint=false and idempotentHint=true, but the description adds valuable context: it opens a visible browser, waits for human input, never types credentials, saves the session to local disk with permissions 0600, and is interactive. This goes beyond annotations and provides security-relevant details that affect agent 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?

  The description is front-loaded with the key action (opens browser, waits for human) and then covers important caveats (never types, session saved, not under dry-run) in a compact set of sentences. Every sentence contributes value, and the structure is scannable.

  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 interactive nature and reliance on human input, the description covers the essential flow: what it does, how it completes (when feed appears, session saved), and constraints (no credential handling, file permissions). It doesn't specify failure handling, but for a login tool this is acceptable and the description is complete for an agent to invoke it 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?

  The tool takes no arguments, and the schema confirms an empty properties object. The description redundantly states 'Takes no arguments.' Since there are no parameters to document, the baseline of 4 applies, and the description adds no misleading information.

  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 purpose: opens a visible Chromium window on LinkedIn's login page and waits for the human to sign in. It specifies the verb (opens, waits), resource (LinkedIn login page), and explicitly distinguishes itself from siblings by emphasizing it requires human interaction and never types credentials.

  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 communicates when to use it: when a human must authenticate interactively, including 2FA/CAPTCHA. It also notes the session is saved and reused by other tools, implying usage before authenticated operations, and states it's unavailable under --dry-run. However, it doesn't explicitly name alternatives like linkedin_session_status for checking login status, or mention 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.

• linkedin_send_messageA
  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), the description discloses the two-call handshake, that confirm:true is required to actually send, that it counts against a daily messages cap, the dry-run behavior, and that InMail is never attempted. It also mentions the refusal condition. This is rich, actionable behavioral context that adds real value over the annotations.

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

  Conciseness5/5

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

  The description is dense yet efficient, front-loading the core action and then packing critical details (handshake, restrictions, dry-run) without fluff. Every sentence adds necessary information and the structure guides the reader through the flow logically.

  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 (two-call handshake, degree restrictions, dry-run mode), the description covers all essential aspects an agent needs to call it correctly: return of previewToken, refusal with not_connected, daily cap effect, and dry-run behavior. Without an output schema, it adequately explains the expected response shape.

  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 0% schema description coverage, the description carries the burden of explaining parameters. It explains profileUrlOrConversationId (URL/slug or thread ID), confirm (triggers the send in the second call), and previewToken (returned from the first call). It does not explicitly describe the text parameter, but its meaning is obvious as the message body. This nearly compensates for the lack of schema descriptions.

  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 purpose: 'Sends one direct message' and specifies the two distinct target types (1st-degree connection via profile URL/slug, or an existing conversation via thread ID). This unambiguously distinguishes it from siblings like linkedin_create_post or linkedin_send_connection_request.

  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 usage conditions: only 1st-degree connections, refusal with not_connected, and the two-call handshake required for sending. It explains when the tool will refuse and how confirmation works, though it does not explicitly name an alternative tool for other messaging scenarios. Still, the context is strong enough for an agent to decide when to use it.

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

• linkedin_session_statusA
  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 reveals that it 'loads the feed once to check', which implies a network call despite being read-only. It also assures that 'no cookie or session content is ever returned' and clarifies the --dry-run behavior. These details give the agent important expectations beyond 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 two sentences with no waste. The core purpose is front-loaded, followed by behavioral notes and the no-arguments statement. Every clause adds necessary 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?

  For a zero-parameter, no-output-schema tool, the description fully covers what the tool does, how it checks (loads feed), what it returns (status, save time, verification time), and edge case behavior (dry-run). Nothing an agent needs to invoke it correctly is missing.

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

  Parameters4/5

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

  The tool has zero parameters, and the description explicitly states 'Takes no arguments.' This meets the baseline for no-parameter tools, and the statement removes any ambiguity, so a 4 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 states a specific verb ('reports') and resource ('saved LinkedIn session'), and specifies the exact outputs: whether it still works, when saved, last verified. This clearly differentiates it from action-oriented siblings like linkedin_login or linkedin_create_post.

  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 makes it clear this is a status check tool by stating what it reports. While no explicit alternatives or when-not-to-use are mentioned, the context of sibling tool names (all performing actions) makes the intended usage obvious. A slight deduction for not explicitly saying 'use before other LinkedIn operations'.

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

• linkedin_apply_to_jobA
  4.8/5.0
  Behavior4/5

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

  Annotations indicate readOnlyHint=false, destructiveHint=true, idempotentHint=false, openWorldHint=true. The description goes beyond annotations by disclosing the two-phase confirmation requirement, the refusal reasons, and that it never submits partial applications. It also clarifies that resumePath is never uploaded anywhere except LinkedIn. While it doesn't detail every failure mode, it covers essential behavioral traits beyond the annotations.

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

  Conciseness5/5

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

  The description is comprehensive yet well-structured. It front-loads the core action, then details each parameter, then explains the confirmation flow and failure modes. Every sentence adds value, and it's written in a logical order. It's long but justified given the complexity.

  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 (5 params, nested objects, validation logic) and lack of output schema, the description provides complete guidance: parameter formats, confirmation flow, refusal cases, failure handling, and quota impact. No critical information missing for an agent to call it correctly.

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

  Parameters5/5

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

  Schema coverage is 0%, so the description must fully explain each parameter. It does: jobId formats, resumePath optionality and accepted file types, answers matching semantics (case-insensitive, substring), confirm flag behavior, and previewToken usage. This adds significant meaning beyond the bare schema properties.

  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 purpose: submitting a LinkedIn Easy Apply application. It specifies the verb (submits), resource (LinkedIn Easy Apply application), and key details like jobId formats and confirmation requirement. It distinguishes itself from siblings like linkedin_search_jobs by focusing on the application action.

  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 explains when to use: for Easy Apply postings, and when NOT to use: external applications and non-Easy Apply postings are refused. It also clarifies the confirm flag flow (preview vs submit). Mentions alternatives implicitly (handling external URL manually). Provides clear context for when this tool is appropriate versus others.

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

• linkedin_scrape_feedA
  4.8/5.0
  Behavior5/5

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

  The description substantially exceeds the annotations' coverage. It discloses the read-only nature (never likes/comments/reposts, no confirm, no quota), explains the lazy-loading with human-like pauses and the time implication for large counts, de-duplication by permalink, skipping of sponsored cards, and that fewer-than-requested results is normal. This rich behavioral detail helps the agent anticipate side effects and performance without contradicting the annotations.

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

  Conciseness5/5

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

  Though longer than average, every sentence earns its place. The opening sentence states the core purpose, then parameter semantics, behavioral safety, de-dup/skipping logic, and the normal-result caveat are each addressed succinctly. The structure is front-loaded and contains no filler or repetition.

  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 tool with one parameter and no output schema, the description covers everything an agent needs: what it returns, how it behaves, potential pitfalls (time, fewer results), and safety guarantees. There are no meaningful gaps that would prevent correct invocation.

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

  Parameters5/5

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

  Schema description coverage is 0%, so the description must fully compensate. It does: it explains that count is how many posts to return, defaults to 10, has a max of 50, and that a larger count involves scrolling and can take a minute or more. This goes far beyond the schema's bare integer min/max, giving the agent the context needed to choose an appropriate value.

  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 states a specific verb ('reads') and resource ('signed-in member's home feed'), and enumerates the exact output fields (author name/headline, post text, permalink, reaction/comment counts, timestamp). This clearly distinguishes it from sibling tools like linkedin_scrape_profile, which targets profiles, and linkedin_create_post, which writes.

  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 unambiguously communicates that this tool is for reading the home feed, which sets it apart from siblings (profiles, jobs, messaging). However, it does not explicitly name alternative tools or state conditions when this tool should be avoided (e.g., 'if you need profile data, use linkedin_scrape_profile'), leaving that inference to the agent.

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

• linkedin_scrape_profileA
  4.8/5.0
  Behavior5/5

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

  Even though the annotations already declare readOnlyHint and idempotentHint, the description adds substantial behavioral detail: it consumes no daily quota, only clicks the '…see more' expander, returns null/[] for hidden sections, and specifies the sole error condition (selector_not_found). It also clarifies connectionDegree semantics and result caps, going well beyond the annotations.

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

  Conciseness5/5

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

  The description is long but densely informative, with the core purpose and return fields front-loaded followed by parameter formats, behavioral notes, and edge cases. Every sentence adds operational value; no filler or tautology is present.

  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 one optional parameter, strong annotations, no output schema, and no nested objects, this description covers all necessary invocation details: input formats, return fields, error behavior, and limits. An agent can correctly select and call this tool without needing additional documentation.

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

  Parameters5/5

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

  The schema only defines profileUrl as a string with no description, giving 0% coverage. The description fully compensates by explaining all accepted input formats, the default when omitted, and what the tool reads from that URL. This gives an agent everything needed to pass the parameter correctly.

  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 states a specific verb and resource: 'Reads a LinkedIn member profile and returns it as structured data,' and enumerates the exact fields returned. It is clearly distinguishable from sibling tools like linkedin_scrape_feed and linkedin_search_jobs, which target different data sources.

  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 gives concrete usage context: accepted profileUrl formats (full URL, path, bare slug) and the behavior when omitted (reads the signed-in member's own profile). It does not explicitly name alternatives or when not to use this tool, but the profile-scoped purpose and sibling names make the choice reasonably clear.

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

• linkedin_search_jobsA
  4.8/5.0
  Behavior5/5

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

  Annotations already mark read-only and idempotent, but description adds substantial behavior: lazy-loading with randomized scroll pauses, scoring and skipping of cards without job IDs (counted in `skipped`), no confirm or daily quota, and the filter precedence rule (filters wins if both present). This exceeds annotation coverage and is genuinely useful.

  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?

  Although lengthy, every sentence adds value: purpose, parameters, behavior, and downstream workflow. Front-loads the core action and returns, then details. No filler; structure is logical.

  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 complex tool with 8 params, nested objects, and no output schema, the description covers return fields, skipped count, lazy-loading performance, and the apply workflow. It also clarifies default/max count and filter precedence. No critical missing information for an agent to call it correctly.

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

  Parameters5/5

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

  Schema description coverage is 0%, yet the description compensates by explaining each major parameter: required keywords, free-text location examples, all four facets with exact enum values (datePosted and experienceLevel), count default/max, and the nested-versus-top-level filter hierarchy. This is more than the schema provides.

  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 verb+resource: 'Searches LinkedIn job postings' and explicitly lists returned fields (job id, title, company, location, Easy Apply support, URL). Distinguishes from siblings because no other sibling handles job search; it's the obvious entry point for discovery.

  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 clear context: explains how to use filters and count, and explicitly routes results to linkedin_apply_to_job for Easy Apply postings. Does not explicitly state exclusions (e.g., when not to use) but the workflow guidance is strong. Lacks a direct 'use instead of X' but that's not needed given sibling separation.

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

• linkedin_send_connection_requestA
  5/5.0
  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) by detailing concrete side effects: no click or send without confirm, quota consumption, note truncation limits per account type, and refusal before quota is spent. This level of transparency is exemplary.

  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 dense but every sentence adds critical information. It is front-loaded with the core purpose and handshake, then layers on parameter details, refusal conditions, and quotas. 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?

  Given the tool's complexity (two-call handshake, refusals, dry-run, quotas, note limits), the description covers every operational aspect. Without an output schema, it explains what the preview returns and what confirm does. An agent can safely and correctly invoke this tool without further context.

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

  Parameters5/5

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

  Schema coverage is 0% (no descriptions in the schema), so the description carries full burden. It explains profileUrl accepts slug or URL, note has 300-char cap (200 for free accounts with silent truncation), confirm triggers the actual send, and previewToken is part of the handshake. All four parameters are clarified.

  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 sends a connection invitation to a member, optionally with a note, and explicitly describes the two-call handshake. It distinguishes itself from siblings like send_message or create_post by focusing on the connection request action and its preview/confirm flow.

  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 usage instructions: the two-call pattern (preview then confirm), when requests are refused (already 1st-degree, pending invite, no Connect control), and the dry-run behavior. It clarifies when not to call with confirm:true immediately, covering usage conditions thoroughly.

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