Server Quality Checklist

Tool Scores

• taxme_list_returnsB
  3.2/5.0
  Behavior3/5

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

  With no annotations, the description carries the full burden of behavioral disclosure. It explicitly discloses that 'unparsable' status may indicate a maintenance/error page or unreadable rows, and warns against concluding 'nothing to file' from such a page. This is valuable, but it doesn't cover other possible behaviors like pagination, return format, 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.

  Conciseness3/5

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

  The description is a single, run-on sentence packing multiple ideas. It front-loads the core purpose ('Tax returns with status') but then adds a lengthy, awkwardly structured clarification. Splitting into two sentences would improve readability without losing information.

  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 the tool has no parameters, no output schema, and no annotations, the description covers the critical edge case of empty vs. unparsable. However, it omits useful context such as possible status values, result ordering, or any requirements. It is adequate but not exhaustive.

  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, so there is nothing for the description to add beyond the schema. The baseline score of 4 applies as the schema fully covers parameter information.

  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 'Tax returns with status' identifies the resource (tax returns) and the information provided (status), clearly implying a listing operation. The additional nuance about empty vs. unparsable results clarifies the tool's output semantics. However, it lacks an explicit verb like 'list' and does not differentiate from sibling tools such as taxme_status.

  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 no guidance on when to use this tool versus alternatives. It does not mention prerequisites like being logged in, nor does it contrast with sibling tools such as taxme_status or taxme_open_return. The entire focus is on interpreting the output rather than usage context.

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

• taxme_menuA
  3.5/5.0
  Behavior3/5

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

  With no annotations, the description carries the responsibility for behavioral disclosure. It usefully explains that a non-return page yields an error naming the browser location rather than an empty list. However, it does not state whether the operation is read-only or how the successful response is structured, leaving some transparency gaps.

  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, consisting of two short sentences. The core purpose is front-loaded, and the second sentence adds a valuable edge case without wasted words. It is appropriately sized for a zero-parameter tool.

  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 tool with no parameters and no output schema, the description covers the essential output (sections and status) and an important error condition. It could be more detailed about the exact response format or possible status values, but the essentials are sufficient given the tool's simplicity and the surrounding sibling 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?

  The tool has zero parameters, so the description does not need to explain parameter meanings. The baseline score of 4 for zero-parameter tools applies here.

  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 identifies the resource as 'left-menu sections of the open return' and mentions their status, which is specific and distinguishes it from sibling tools like taxme_status. However, it lacks an explicit verb like 'get' or 'list', so it is clear but not perfectly framed.

  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 no explicit guidance on when to use this tool versus alternatives. It implies it operates on the 'open return' but does not state prerequisites or contrast with other tools. No exclusions or alternative tool references are given.

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

• taxme_snapshotA
  3.5/5.0
  Behavior3/5

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

  With no annotations, the description carries the burden of behavioral disclosure. It states the primary output (breadcrumb/url) and the optional screenshot behavior, but does not mention side effects, error cases, or what happens if no page is loaded. The behavior is largely read-only, but this is not explicitly confirmed.

  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 sentence with a semicolon, conveying both the default behavior and the optional flag. Every word is purposeful with no redundancy or filler.

  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 tool with only one optional parameter and no output schema, the description provides the essential information: what information is returned and how to request a screenshot. However, 'breadcrumb/url' and 'PNG path' are ambiguous (e.g., is it a local file path or a URL?), leaving minor gaps.

  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 single boolean parameter 'screenshot' is described in the description ('set screenshot:true for a PNG path'), adding meaning beyond the bare schema. The schema covers the parameter type, but the description clarifies its functional effect, which is helpful given the low schema description coverage.

  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 that the tool provides the current page's breadcrumb/url and optionally a PNG screenshot path. It effectively distinguishes itself from sibling tools by focusing on page-state retrieval, though it lacks an explicit verb like 'get' or 'return'.

  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 is given on when to use this tool versus alternatives. It does not mention that taxme_status or other navigation tools might be more appropriate for certain scenarios, nor does it state any prerequisites or exclusions.

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

• taxme_fillA
  3.9/5.0
  Behavior5/5

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

  With no annotations provided, the description fully carries the transparency burden. It discloses detailed behavior: how different value types are interpreted (text, radio, checkbox), what gets refused (invalid values, disabled fields), and the verification process (read-back and ok:false if mismatch). This is rich behavioral context that goes beyond basic expectations.

  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 description is a single, lengthy run-on sentence that packs in a lot of information but is poorly structured. It could be broken into clearer sentences or bullet points for readability. Every piece of information is relevant, but the format hurts comprehensibility.

  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 the tool (multi-field form filling with target resolution, value mapping, refusal conditions, and verification), the description covers all major aspects. It does not describe the full reply structure but does mention ok:false. It is complete enough for an agent to invoke the tool correctly with no output schema.

  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 a generic 'values' array with 'target' and 'value', but the description adds crucial semantic meaning: target can be a field id or label/context substring, and value must be text/number/boolean with type-specific handling. Since schema coverage is 0%, this description fully compensates and even exceeds what the schema provides.

  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 verb+resource: 'Set fields on the current page.' It is specific enough to distinguish from other tools like taxme_click or taxme_menu, though it does not explicitly name any alternatives. The meaning is unambiguous and matches the tool name.

  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 does not provide guidance on when to use this tool versus alternatives, nor does it mention prerequisites or exclusions. It only implies usage on a page with fields, but there is no explicit 'use this when...' or comparison with sibling tools.

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

• taxme_resultsA
  4.1/5.0
  Behavior5/5

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

  With no annotations, the description carries the full burden. It reveals several key behaviors: the requirement to have the section open via breadcrumb, the error behavior when the portal refuses (returns the actual page name), the rule that the calculation is only read from a line containing HEADS and only when that line is a heading element, and the handling of expired sessions (reports no return open). This is rich behavioral disclosure.

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

  Conciseness2/5

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

  The description is a single dense paragraph with many nested clauses. While the first sentence is clear, the rest is a long, convoluted explanation of edge cases that could be broken up or summarized. It is not concise or easily scannable.

  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?

  The description covers the main purpose, preconditions, error cases, and the specific rule for identifying the calculation line. However, it does not explicitly describe the return value format beyond implying it reads an amount. Given the absence of an output schema, this is a minor gap, but overall the tool is well-contextualized.

  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, so there is nothing to document beyond the schema. The baseline for 0-param tools is 4, and the description doesn't need to add parameter details.

  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 statement: 'Read the Ergebnisse / Steuerberechnung of the open return.' This specifies the verb (read), the resource (Ergebnisse/Steuerberechnung), and the scope (open return). It distinguishes from sibling tools like taxme_get_fields and taxme_snapshot by focusing on the calculation section.

  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 the tool is for reading the tax calculation once the section is open, but it does not explicitly state when to use it versus alternatives. It mentions preconditions (reaching the section) but does not name any alternative tools or exclusion criteria. Thus, usage is implied rather than explicitly guided.

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

• taxme_account_statementA
  4.2/5.0
  Behavior5/5

  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 and does so exceptionally well. It explains critical edge cases: amounts are only reported under the year the statement assigns, and if none can be assigned, the status is 'unparsable' rather than an empty list. It also clarifies that non-statement pages (maintenance, error) are 'unparsable', preventing misleading empty-list interpretations.

  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 three sentences, with the first stating purpose and the next two providing essential edge-case behavior. Every sentence adds value, though it is slightly more verbose than strictly necessary. The information is front-loaded and no unrelated details are included.

  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 parameterless tool with no output schema, the description is remarkably complete. It defines the primary output (tax amounts per year), the meaning of 'unparsable' status, and the distinction between empty list and unparsable. This is sufficient for an agent to correctly interpret results, especially given the lack of structured output annotations.

  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, so there is nothing for the description to add beyond the empty input schema. Baseline for zero parameters is 4, and the description does not need to explain parameter 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 first sentence clearly states the tool's function: 'Open tax amounts (CHF) per tax year.' This is a specific verb+resource pairing that distinguishes it from sibling tools like listing returns or getting fields. The subsequent sentences further clarify scope by describing how amounts are attributed.

  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 no guidance on when to use this tool versus alternatives. It does not mention sibling tools, exclusion conditions, or prerequisites. The only implied usage is 'when you need tax amounts per year,' which is tautological and not helpful for tool selection.

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

• taxme_open_returnA
  4.3/5.0
  Behavior4/5

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

  With no annotations provided, the description thoroughly explains expected statuses (ok, login_required, not_open, wrong_year) and the verification against the requested year. It also notes handling of the edit popup tab, adding useful 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?

  Two sentences only, front-loaded with purpose and outcome. Every word adds value; no fluff or repetition of schema fields.

  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 single-parameter tool with no output schema, the description covers purpose, return value (menu sections), and all possible status outcomes, making it fully self-contained for an agent.

  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 has no descriptions, so the description carries the burden. It identifies 'year' as the tax return year and explains it is used for page verification, but does not specify format or constraints beyond the schema's type.

  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 it opens a tax return for editing and returns menu sections, using a specific verb and resource. It is distinct from sibling tools like taxme_status or taxme_list_returns.

  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 for when to use (to open a tax return for editing), but does not explicitly name alternative tools or exclusion conditions. The implied use case is evident from the description.

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

• taxme_get_fieldsA
  4.4/5.0
  Behavior5/5

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

  With no annotations, the description carries full burden. It discloses truncation behavior (limit, default 60), that the reply reports omitted fields, the meaning of 'locked', and that taxme_fill still sees all fields. This goes well beyond a simple 'list fields' statement.

  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 a single information-dense and front-loaded sentence, but it becomes a bit run-on with parentheticals and semicolons. Every clause earns its place, yet splitting into two sentences would improve readability.

  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 no output schema, the description fully covers what the tool returns (list of fields with specific attributes), the limit behavior, the locked concept, and the interaction with taxme_fill. This is complete for a low-complexity read-only tool with one parameter.

  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% for the single 'limit' parameter, so baseline is 3. The description adds behavioral semantics: limit causes truncation and the reply states how many were left out, which the schema alone does not convey.

  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 'List interactive fields on the current page' – a specific verb and resource. It enumerates the field attributes (id, type, value, label, context, radio group name, locked) and distinguishes itself from siblings by referencing taxme_fill behavior.

  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 (use this to see fields on the current page) but no explicit when-to-use or alternatives are given. The mention that taxme_fill resolves against every field hints at a workflow but does not state 'use this before filling' or contrast with sibling tools like taxme_snapshot.

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

• taxme_statusA
  4.5/5.0
  Behavior4/5

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

  No annotations are provided, so the description carries full burden. It discloses the side effect of refreshing the cached session and explicitly indicates the possible return values. It doesn't mention error states or authentication details, but for a simple status check this is adequate.

  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 exactly two sentences, front-loaded with the main purpose, and includes the key usage instruction. Every word 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?

  Given the zero-parameter tool and no output schema, the description fully covers what the tool does, what it returns, when to call it, and a key side effect. It is complete for an AI 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?

  The tool has zero parameters and the input schema is empty, so the baseline is 4. There is no parameter information needed; the description appropriately focuses on behavior and usage.

  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 checks the BE-Login/TaxMe session status and returns 'ok' or 'login_required'. This specific verb and outcome, plus the instruction to call it before anything else, distinguishes it from siblings like taxme_login and taxme_list_returns.

  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 phrase 'Call this before anything else' provides explicit when-to-use guidance. While it doesn't name alternative tools or specify when not to use it, it clearly positions this as the initial health check, which is sufficient context given the sibling list.

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

• taxme_goto_sectionA
  4.7/5.0
  Behavior5/5

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

  Without annotations, the description carries full responsibility for behavioral disclosure, and it excels. It details substring resolution via breadcrumb paths, error handling for form validation and expired sessions, and the truncation behavior ('cut at 60... with truncated/total'). This goes far beyond a basic operation description.

  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 and lengthy, but it is front-loaded with the core action in the first sentence. Every subsequent sentence adds necessary nuance for a complex matching algorithm and error scenarios. It is appropriately sized for the complexity, though not as crisp as a simpler tool's description.

  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 single parameter, no annotations, and no output schema, the description covers all essential aspects: operation, output format (fields with truncation), error conditions (form errors, expired session), and resolution logic. It also references taxme_get_fields to clarify output behavior, making it effectively complete for an agent.

  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 only parameter 'name' receives extensive semantic explanation: it is a substring that must match a menu entry as determined by breadcrumb resolution, with examples ('Wertschriften' at front, 'Einkünfte' at end). This adds tremendous meaning beyond the bare schema string type, especially given 0% schema description coverage.

  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 primary function: 'Click a menu section by name (substring) in the open return; returns the fields on that page.' This specific verb+resource combination distinguishes it from siblings like taxme_menu (which lists sections) and taxme_get_fields (which fetches fields without navigation).

  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 implies usage context by specifying 'in the open return' and notes a critical precondition: 'TaxMe refuses to open a section while the form still has errors.' It does not explicitly name alternative tools or when-not-to-use scenarios, but the context is clear enough for an agent to select this tool for section navigation.

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

• taxme_loginA
  4.7/5.0
  Behavior5/5

  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 goes beyond a simple statement: it explains the visible window, the up-to-8-minute wait, the post-login status check, the reason for asking the portal, and that nothing is cached. This is rich, honest transparency about the tool's behavior and limitations.

  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 yet informative. It front-loads the core action, then provides essential caveats in two follow-up sentences. Every sentence adds value: the visible window, the timeout, the meaning of 'ok', and the caching behavior. No wasted words.

  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 is remarkably complete. It explains the tool's purpose, its blocking behavior, expected outcomes ('ok', 'login_required'), and a key nuance about the portal address. There is no missing information that would prevent an agent from using or interpreting 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?

  The tool has zero parameters, so the baseline is 4 according to the rubric. The description does not need to explain parameters, but it does add contextual meaning about the login flow, which indirectly clarifies what the tool does despite having no inputs.

  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 action ('Open a visible window for the SwissID/AGOV login') and distinguishes it from sibling tools like taxme_status or taxme_get_fields by focusing on the authentication step. The verb+resource structure is specific and unambiguous.

  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 implies when to use (at the start of a session requiring login) and provides critical guidance on interpreting the result ('Only status "ok" means the login went through'). It does not explicitly name alternatives or exclusions, but no direct alternative exists among the siblings, and the guidance is sufficient for correct usage.

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

• taxme_submit_returnA
  4.7/5.0
  Behavior5/5

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

  With no annotations, the description carries full burden and does so impressively. It discloses irreversibility, dry-run behavior when confirm is false, the would_click/not_clicked distinction, and every refusal case including expired sessions. This goes far beyond basic operation and fully informs the agent of all safety behaviors.

  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 a single verbose paragraph, but every sentence communicates a distinct edge case or behavioral rule, earning its place. It is longer than ideal, but the high stakes justify thoroughness. Minor deduction for lack of structural formatting like bullet points.

  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 complexity of an irreversible submission, no annotations, and no output schema, the description is exceptionally complete. It covers preconditions, response outcomes (dry-run vs actual click), and all refusal scenarios, leaving the agent well-equipped to invoke 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?

  The schema only defines a boolean 'confirm' with no description. The description explains that confirm:true is required for actual submission, and false triggers a dry-run, adding meaning that is absent from the schema. Since schema coverage is 0%, this is essential and well-handled.

  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 'DANGER: final submission' and 'Abschluss → Steuererklärung einreichen', clearly indicating this tool performs the final tax return submission. It is distinctly different from sibling tools by emphasizing irreversibility and the submit button, leaving no ambiguity about what action it performs.

  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 explicit preconditions: the user must be on the Abschluss page and confirm must be true. It also details when the tool refuses to act (already submitted, page unreadable, no menu). However, it does not explicitly name alternative sibling tools for other contexts, so it lacks direct comparative guidance.

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

• taxme_clickA
  4.8/5.0
  Behavior5/5

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

  With no annotations, the description carries the full burden and exceeds expectations. It discloses the exact match preference, substring fallback, reporting of the actually pressed button, and the refusal of empty labels with reasoning. This is rich behavioral context beyond the bare minimum.

  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 purpose, then adds necessary details and examples without redundancy. Every sentence contributes value—examples clarify usage, and the empty-label warning prevents misuse. It is concise yet comprehensive.

  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 single-parameter tool with no output schema or annotations, the description is complete. It covers what the tool does, how it behaves in ambiguous cases, and how errors are handled. No relevant aspect is missing, making it fully self-contained 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.

  Parameters5/5

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

  The schema only provides a 'label' string with no description, so the description fully compensates. It explains that the label is the visible text, gives examples, and details the matching semantics (exact vs substring) and the empty-label edge case, making the parameter meaning unambiguous.

  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 action: 'Click a button/link by visible text', which is a specific verb and resource. It provides concrete examples of labels and distinguishes itself from sibling tools like taxme_fill by focusing on clicking existing elements, not filling or navigating.

  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 clear context for when to use the tool (clicking any button/link by visible text) and includes examples of typical labels. It does not explicitly name alternatives or exclusions, but the scope is well-defined and easily inferred from the examples and the nature of the action.

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