MachineWitness

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

Annotations already declare readOnlyHint=true, idempotentHint=true and destructiveHint=false, so safety is covered structurally; the description adds the closed-world state vocabulary (awaiting payment, queued, capturing, first delivery sent, complete, plus per-witness waiting/capturing/captured/failed/refused) and an explicit negative scope ('never returns the e-mail address, the addresses ordered or anything else about the customer'), which is genuinely useful context. It stops short of stating behavior on an unknown reference or stale record.

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 compact sentences, front-loaded with the resource and its state vocabulary, ending with the privacy boundary. The two enumerations are information-dense rather than padding, though the second ('Per witness: …') interrupts the flow slightly.

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 correctly takes on the burden of describing return values by enumerating both record-level and per-witness states, and it bounds what is not returned. It omits error/invalid-reference behavior and any indication of whether status can change over time (polling expectations), leaving a small gap for a point-in-time status tool.

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% and the single parameter carries its own description with an example, so the schema does the heavy lifting. The description restates the reference prefix (MW-K-…) but adds no format, validation, or sensitivity 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?

The description states a specific resource (a point-in-time record) and the information returned (its status, keyed by reference), and enumerates the possible states, so the agent knows exactly what it gets back. It does not, however, distinguish itself from siblings such as check_coverage or order_capture, so an agent still has to infer the boundary. Clear purpose, no sibling differentiation.

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?

Usage is only implied: 'by its reference (MW-K-…)' signals a lookup keyed on an existing order reference, and the enumerated states imply a polling/monitoring use. There is no explicit when-to-use-this versus check_coverage or order_capture, and no stated preconditions. Minimum viable guidance.

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 readOnlyHint, idempotentHint, non-destructive, and closed-world, so the safety profile is covered. The description adds real value beyond that: it states the return is metadata only, never file contents, never an assessment of the domain or operator, and pre-empts misreading 'not observed' as a judgment about the domain. It does not mention auth or rate limits.

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 what the tool returns, followed by output-scope limits and a caveat about the meaning of a negative result. Every sentence carries information, though the opening sentence is long and packs four distinct facts into one clause chain.

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?

There is no output schema, so the description carries the burden of describing the return — and it does, listing sealed day, requested files, and cadence. Combined with the explicit statement that no file contents or judgments are returned, an agent has enough to call it correctly; only error/edge behavior (invalid domain, subdomain) is unaddressed.

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?

Only one parameter exists and schema description coverage is 100%, including the note that a pasted URL is accepted. The description adds no additional semantics about the domain argument (normalization, subdomain handling, IDN), so the schema does the work and the baseline of 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?

The description names a specific verb and resource — answering whether this archive observes a given domain — and enumerates the dimensions of the answer (sealed day, requested machine-readable files, cadence). It is unambiguous about what the tool does, but it never references or contrasts with siblings like capture_status or suggest_domain, so an agent gets no explicit routing signal.

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?

Usage is implied: this is the lookup to run when you want to know if a domain is in the archive. There is no explicit when-to-use/when-not guidance and no named alternative (e.g., suggest_domain for unobserved domains, order_capture to add one), so the agent must infer routing from sibling names alone.

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 only declare readOnlyHint=false, openWorldHint=true, idempotentHint=false, destructiveHint=false. The description goes well beyond them: no charge at call time, per-address refusal with named reason, delivery by e-mail plus a second e-mail after the night's seal, and the fact that no purpose is collected. That is exactly the behavioral context an agent needs and cannot get from structured fields.

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 purpose is front-loaded in the first sentence, but the paragraph is a single run-on stream with redundancy (address count/domain repeat the schema, and the fees URL is tacked on). The behavioral detail earns its place, but tighter sentence structure and removal of schema echoes would improve it.

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 burden and largely does so: a payment link page is returned, refused addresses are named with reasons, and results arrive by e-mail. It does not specify the exact return shape or error object, leaving a small gap for a mutating tool.

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 the schema already documents email, language, timezone and addresses. The description's 'up to twenty public addresses, all on the same domain' and 'only addresses are asked for, never a purpose' largely restate the schema rather than adding format or constraint detail, so the 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?

The description states a specific verb and resource: 'Create a payment link for a point-in-time record' of named public addresses, with clear scope (up to twenty, one domain). It is not a tautology of the name/title. It does not explicitly differentiate itself from siblings like check_coverage, whose pre-validation role it silently absorbs, so it stops 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.

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

Gives strong usage context: the call charges nothing, validation happens automatically before link creation, and the capture only starts after a person pays on the returned page. However it never names an alternative tool or states when NOT to use this one, so the routing guidance against check_coverage/capture_status is left implicit.

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 annotations: it discloses that the call creates no obligation, no timeline, and no assurance of addition, that the response is always 'received' with 'guarantee: none', and that a human reads the reason. This is exactly the kind of non-obvious behavioral context the annotations cannot convey.

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?

Four sentences, front-loaded with the core action, then the request-not-instruction framing, then the outcome. Every sentence carries distinct information with no 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?

Despite having no output schema, the description fully describes the return ('received', 'guarantee: none') and the downstream process, so an agent knows both what to send and what to expect.

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 baseline is 3, but the description adds meaning the schema lacks: the reason is required and is read by a person, framing it as human-facing prose rather than a machine field.

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 ('suggest that a domain be observed') and frames the action as a non-binding request. It implicitly separates itself from order_capture, but never names the sibling or explicitly contrasts the two.

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?

Explains what happens after the suggestion (criteria, operator decision) but never says when an agent should prefer this over order_capture or the other siblings. Usage is implied rather than stated.

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