DokuTrak

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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuinely new context: 'Received does not mean approved', 'This tool sends no reminder', and that file reading or approval decisions must happen in DokuTrak — useful domain semantics an agent would otherwise have to guess.

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?

Front-loaded with the return payload, then lookup mechanics, then disambiguation, then two scoping caveats. Five short sentences with no filler; slightly breathless but every sentence carries information.

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 steps in by enumerating what is read (receipt, approval status, deadlines, reminder progress) and where the remaining actions live (DokuTrak). This is sufficient for an agent to call and interpret the tool correctly, missing only explicit sibling routing.

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 0%, so the description must compensate, and it does: it explains that 'search' matches title, Client name or email and that 'request_id' is the precise alternative. It does not mention format constraints (UUID) or the either/or exclusivity, but the semantic meaning of both params is now clear.

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?

The description opens with a concrete verb+resource: read a document request's checklist receipt, Human Approval status, deadlines and reminder progress. It is clear what the tool returns, but it never names or contrasts itself against siblings like prepare_request or review_request, so the differentiation is left to inference.

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?

It states exactly how to identify the target ('Give request_id or search the title, Client name or email') and prescribes a disambiguation flow when multiple requests match. It also implicitly excludes reminder sending, though it does not explicitly name send_request as the alternative for that action.

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/non-destructive, so the safety profile is covered. The description adds real value beyond them: no request is created, nothing is sent, the preview must survive the connection step and be forwarded unchanged (an idempotency-preserving constraint), and invalid details must be corrected first.

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 short sentences, front-loaded with the scope disclaimer before the workflow instruction. The closing 'Missing or invalid details must be corrected' is the vaguest line and restates an obvious validation expectation, but nothing is padded.

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 usefully explains the preview that comes back and how to carry it forward, and it clarifies the non-sending semantics. However, for a nested-object tool at 0% schema coverage it stops short of describing the document/checklist item structure or any field constraints an agent must satisfy.

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 0% for a single nested object with many subfields, so the description carries the burden. It names recipient, checklist, deadline and optional message, which maps to recipientName/Email, documents, deadline and message, but leaves title, recipientLanguage, and all document subfields (doc_type, required, coverage_days, maxItems=10) unexplained.

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?

Specific verb+resource (prepare/edit a request preview) plus an explicit boundary: 'No request is created and nothing is sent.' It names the sibling it hands off to (review_request), so an agent can place it in the flow without opening any other schema.

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?

States the ordering constraint clearly: prepare 'before connecting,' then keep the returned preview through connection and pass it unchanged to review_request. It does not explicitly distinguish itself from get_request or send_request as alternatives, but the sequencing context is unambiguous.

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?

Beyond the annotations, it discloses the most important behavioral fact: review creates no request and sends no email, and an explicit human confirmation in DokuTrak is required. It also reinforces the idempotency contract declared by idempotentHint (reuse the operation_id) and explains the retry-vs-new-operation boundary. It does not say what the response contains or how long a review stays pending.

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?

Five short sentences, each carrying a distinct rule, with the primary action front-loaded and the negative guarantee ('No request or email is created') placed last for emphasis. No filler, though the opening 'After connection' phrasing is slightly opaque.

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 and no annotation about return values, the description does well to mention reviewUrl, which is otherwise undiscoverable. But for a tool with a deeply nested preview object at 0% schema description coverage, the absence of any guidance on what a valid preview contains leaves a meaningful 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 0%, so the description carries the full burden, and it does add semantics: the preview must be 'unchanged' and operation_id must be a stable UUID reused across retries, which tells the caller these two values are a matched pair. It still says nothing about the nested preview fields (title, message, deadline, documents, recipient*), leaving the caller to infer them from the schema alone.

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?

The description conveys a specific action (submit the unchanged preview for review) and a specific outcome (the Professional reviews and confirms in DokuTrak via reviewUrl), and it explicitly clarifies that no request or email is created. That last clause implicitly distinguishes it from send_request, but neither send_request nor prepare_request is named, so sibling differentiation is indirect rather than explicit.

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?

It gives real operational guidance for retries ('Reuse the operation_id for retries') and for edits ('Changed details require a new operation and confirmation'), and 'After connection' implies sequencing. However, it never states when to choose this tool over prepare_request, send_request, or get_request, so the primary routing decision is left to inference.

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?

Goes well beyond the annotations: it clarifies that retries recover the same request (reinforcing idempotency with operational detail), redefines the 'sent' outcome as provider acceptance rather than inbox delivery, and names the delivery_unknown state with required follow-up. These are exactly the semantic distinctions an agent cannot infer from idempotentHint/openWorldHint alone.

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?

Five short sentences, each carrying distinct operational information (precondition, parameter provenance, retry semantics, success meaning, error handling). The precondition is front-loaded and there is no filler.

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 an irreversible, open-world send operation with no output schema, the description covers the precondition, the success semantics, the ambiguous failure state, and retry behavior. An agent has everything needed to invoke it correctly and interpret the outcome.

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 0%, so the description must compensate: it tells the agent where operation_id and version come from (review_request / the reviewed version), which is the key provenance constraint. It does not explain the 64-character version format or UUID constraint, but the schema's type/format/pattern fields already constrain those values.

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 concrete action on a concrete resource (send the confirmed Document Request version), and routes the agent through the review step by naming reviewUrl and review_request. It does not explicitly contrast itself with get_request or prepare_request, so sibling differentiation is implied rather than spelled out.

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 an explicit precondition ('Send only after the Professional confirms the exact version on reviewUrl') and an explicit prohibition for the failure path ('delivery_unknown requires inspection in DokuTrak; do not start over or resend'). It also closes the alternative route by stating no immediate reminder is available.

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