Digital Help

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

Annotations already declare readOnlyHint, idempotentHint, openWorldHint=false and destructiveHint=false, so the safety profile is covered. The description reinforces this with 'Only computes; stores nothing', which is mildly redundant, and it does not discuss rate staleness beyond naming the 'month the rates were last updated' return field. Useful but not adding much beyond structured data.

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

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

Three sentences, front-loaded with purpose, then return values, then scope limits — nothing is wasted. The opening sentence is dense with clauses but still readable, keeping it short of a 5.

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

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

With no output schema, the description deliberately enumerates the return payload (gross, net, income tax, contributions, employer cost, effective rate, rate update month, link), which is exactly what an agent needs to reason about results. Parameters are fully schema-documented and annotations cover the safety profile, leaving no material gap.

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

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

Schema description coverage is 100%, so all five parameters (amount, locale, period, region, direction) are already documented with units and enum meanings. The description restates the direction modes in prose but adds no syntax, currency, or default detail beyond the schema — baseline 3 applies.

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

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

States a specific verb and resource ('Calculate take-home pay from a gross salary') plus the exact scope (Albania or Kosovo) and the inverse direction (gross needed for a target net). An agent can distinguish this from the unrelated siblings (create_cv, run_tool, search_tools) immediately.

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

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

Explicit when-to-use ('salary, wage or payslip questions about Albania or Kosovo') paired with an explicit when-not ('it doesn't cover other countries'). No sibling alternative is needed because the other tools are unrelated, and the two computation modes are stated up front.

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

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

Adds substantial context beyond the annotations: the draft is stored on Digital Help's server for up to 60 minutes, is deleted on first link open, and each call creates a new link (which concretely explains the non-idempotent behavior flagged by idempotentHint=false). It also discloses the return payload (outline + link), which no output schema provides.

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

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

Three tight sentences: capability + return, usage trigger, side effects. Zero restatement of the schema, and the most decision-relevant facts (editor link, retention window) are front-loaded.

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

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

For a nested, non-idempotent mutation with no output schema, the description covers the critical unknowns: what is returned, where the draft lives, how long it persists, and how to avoid inventing user data. It leaves template/section semantics to the schema, which is only 60% described — the one remaining gap.

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

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

Schema description coverage is 60%, so much is already documented (locale enum, headline, section title, entry fields, date patterns). The description adds only the input-provenance constraint ('include only details the user provided') and confirms the language set; it says nothing about the template enum, section ordering, or the 10-section/30-entry caps, so it does not compensate for the ~40% coverage gap.

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

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

States a specific verb and resource ('Create a CV (resume) from details the user has given') plus the concrete outcome: a link to the Digital Help CV editor and a returned text outline. The editor/PDF-DOCX framing makes it unmistakable against the unrelated siblings (calculate_net_salary, search_tools).

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

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

Gives explicit trigger conditions — 'when the user asks to write, build or format a CV' — and narrows scope to the three supported languages plus the constraint to 'include only details the user provided'. No when-not guidance or named alternative, but there is no comparable sibling to route to, so the omission is minor.

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

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

Annotations already declare non-read-only, non-destructive, and non-idempotent behavior. The description adds important context beyond annotations: calculators only compute, krijo-cv saves a CV draft for 60 minutes behind a one-time link, and unsupported slugs return only a shareable page URL.

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

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

The description is four tight sentences. The core run-and-return behavior is front-loaded, followed by exceptions and side-effect details, with no filler or repetition.

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

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

For a dynamic wrapper with nested input and no output schema, it explains the result (tool output plus page link), fallback URL-only behavior, regional requirements, and CV draft lifetime. This is sufficient for correct invocation.

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

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

Schema coverage is 100%, so the baseline is 3. The description adds cross-tool meaning for slug (from search_tools, callable:true), input (same fields as the corresponding named tool), and region (required for regional tools), though it adds little for locale beyond the schema.

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

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

States a specific verb ('Run') and resource ('a tool by slug'), identifies the slug source as search_tools with callable:true, and names sibling equivalents: paga-neto/calculate_net_salary and krijo-cv/create_cv. This clearly distinguishes run_tool from search_tools and the named tools.

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

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

Explicitly says to use slugs returned by search_tools that are callable:true, notes that paga-neto and krijo-cv mirror the named tools, and states that any other slug only returns a page URL. It also notes regional tools require region, covering when to use it and what happens otherwise.

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

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

Annotations already declare readOnly, idempotent and non-destructive, so 'Read-only; changes nothing' is largely redundant, but the description adds real behavioral value absent from annotations: a hard cap of 8 matches, per-result fields (slug, title, description, page URL, regional), and the regional split between Albania and Kosovo pages. With no output schema, this return-shape disclosure is the main contribution.

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

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

Three tight sentences, front-loaded with the action and scope, then the return contract, then the safety note. The topic list is long but functional; little waste overall.

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

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

With no output schema, the description carries the return contract and does so adequately (result count cap and field list). Minor omissions remain, such as what happens on zero matches or how ranking works, but nothing blocks correct invocation.

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

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

Schema description coverage is 100%, with the query examples ('net salary', 'tvsh', 'car customs') and the sq/en enum already documented. The description's mention of topics and Albanian/English input restates rather than extends the schema, so baseline 3 applies.

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

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

States a specific verb and resource ('Search the 200+ DigitalHelp tools ... by topic') plus an enumerated topic domain, so the tool's function is unambiguous. It also distinguishes itself from siblings by noting that matches runnable on this server name the tool to call, e.g. calculate_net_salary.

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

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

The description makes the use case clear: keyword-based discovery across Albanian/Kosovo topics, with the chaining hint that results may point to an executable tool. It does not explicitly state when NOT to use it (e.g. call a known tool directly) or how it relates to run_tool, but the context is clear enough.

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