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

No annotations exist, so the description carries the full burden, and it does disclose real behavior: missing or empty sections are surfaced under `unreadable` rather than inferred, and `unedited: true` signals template placeholders. That is valuable edge-case and return-field disclosure an agent could not infer from the empty schema. It does not state read-only-ness, staleness/caching, or what happens if handoff.md is absent entirely.

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 the purpose, then two sentences each dedicated to a distinct return-value caveat. No filler, no restatement of the tool name, and every sentence changes how the agent would interpret the result.

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 annotations, the description does the necessary work of explaining what the response contains (three permission tiers, `unreadable`, `unedited`) and how to act on it. It is nearly complete for a zero-parameter lookup; only the absent-file case and the freshness of the read are 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?

There are zero parameters, so the baseline is 4; there is nothing for the description to clarify. The description instead spends its words on output semantics, which is the right trade-off.

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 resource (the user's handoff.md policy) and its content domain: what the agent may do alone, must ask first, and must never do. That is a clear verb-less but unambiguous purpose, distinguishable from siblings like handoff_read or handoff_status. It stops short of explicitly contrasting itself with those siblings, so 4 rather than 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?

Usage is implied — an agent should consult this before acting autonomously — but the description never says when to call it (e.g., at session start, before any write action) nor when not to. It also doesn't name any of the seven sibling tools as an alternative for related lookups.

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?

With no annotations, the description carries the full burden and delivers the most important thing: a prompt-injection warning that reply text is untrusted and must not be acted on without user confirmation. It also discloses ordering (oldest first) and the status vocabulary, though it never states read-only-ness or whether consuming the inbox marks items answered.

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?

Two sentences, both earning their place: the first identifies the resource and ordering, the second delivers the security constraint. The ordering detail and the trust warning are front-loaded rather than buried.

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?

The security framing is unusually complete, but with no output schema the description should sketch the returned shape; 'the text in the reply' hints at a reply field without defining it, and the default filter behavior is unstated. It is adequate rather than complete for a retrieval 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?

There is a single optional enum parameter at 0% schema description coverage, so the description should compensate. It echoes the enum values (new or answered) but does not say what omitting status returns — presumably both — which is exactly the ambiguity an agent needs resolved.

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 concrete resource and scope: notes left by another agent in the desk's inbox folder, returned oldest first with a status. It distinguishes 'inbox' from the outbox side of the handoff family implicitly, though it never names a sibling to contrast with.

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 'what the agent left for Claude' — an agent can infer this is how you pick up peer messages, but there is no explicit when-to-use or when-to-prefer-handoff_read/handoff_wait/handoff_list 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?

With no annotations, the description carries the behavioral burden and does well: it states that it appends a reply heading, marks the note answered, may push when the desk syncs through git, and skips duplicate identical replies. It omits error behavior for missing notes and any permission requirements.

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 compact and front-loaded: it opens with the action, then explains side effects, then defines `file`, then notes duplicate skipping. Every sentence contributes useful 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?

For a two-parameter mutation tool with no annotations or output schema, the description is fairly complete: it explains the main side effects, git push behavior, and duplicate handling. It could still say more about return values or failure modes, but the core operational context is present.

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%, so the description must explain both parameters. It clarifies that `file` is the note's file name in the inbox folder, but `reply` is only implied by 'appends ... reply' and lacks any description of expected content or constraints.

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 gives a specific verb and resource: reply to an inbox note by appending a reply and marking it answered. It is clear what the tool does, though it does not explicitly differentiate itself from siblings like handoff_inbox or handoff_read.

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: use this to answer a note already in the inbox, with `file` naming that note. There is no explicit guidance on when to choose this over handoff_send, handoff_read, or how to first locate the note via handoff_inbox.

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?

With no annotations provided, the description carries the full burden, and it does disclose a meaningful failure mode: if the shared repository is unreachable it answers from a local copy and returns a different shape, { stale, items }. It still omits auth/permission needs, ordering, and any result-size or pagination behaviour, which keeps it short of a 5.

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?

Two dense sentences, front-loaded with the verb, resource and returned fields, then the filter option, then the degradation caveat. Nothing is wasted and no sentence merely restates the name.

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 single-parameter, read-only list tool with no output schema and no annotations, the description covers return fields and the stale-fallback shape adequately. The remaining gaps are the undocumented "info" enum value and the absence of ordering/pagination expectations.

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 does explain the filter is optional and that omitting it lists everything, which the schema does not say. However it advertises only waiting/done/declined while the enum also contains "info", leaving one accepted value unexplained and creating a risk the agent assumes "info" is unusable as a filter.

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 ("List desk items") and even enumerates the returned fields (id, type, title, status, created), so the agent knows exactly what it gets. It does not differentiate from near-neighbours such as handoff_inbox or handoff_read, which likely also surface items.

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 only implies usage through the optional status filter ("optionally only those waiting, done or declined"); it never says when to prefer this over handoff_inbox, handoff_read or handoff_status. No exclusions or prerequisites are given.

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?

With no annotations, the description carries the burden and does real work: it discloses graceful degradation ('stale' note when the shared repository can't be reached) and an explicit untrusted-content safety stance about the other agent's reply. Missing only error/not-found behavior and any permission requirements.

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 read purpose and return fields before the safety caveat; both sentences earn their place. The second sentence is long but the untrusted-content warning justifies the length.

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?

No output schema exists, and the description compensates by enumerating return fields including the conditional 'stale' state; with no annotations it also supplies the security posture. Only failure modes (invalid or missing id) are left uncovered.

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 the single 'id' parameter, so the description should compensate; it only implies that the id selects the desk item, without stating the identifier space or that it must be a positive integer. Baseline 3 given one parameter whose role is at least contextually implied.

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 ('Read') plus a precisely scoped resource ('one desk item') and an enumeration of what comes back: request, reply, status, timestamps. An agent can distinguish this single-item read from the list-shaped siblings (handoff_list, handoff_inbox) without opening any 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?

The 'one item by id' framing implies the calling condition, but the description never states when to prefer this over handoff_list or handoff_inbox, nor what to do if the id is unknown. Usage is inferable rather than stated.

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?

With no annotations, the description carries the full load and does well: it discloses that the handoff lands as a note in the desk folder, gets pushed to a shared repo on git sync, is emailed when a doorbell is configured, that the agent replies inside the note, and what to do when ring.rung is false. It also warns against including secrets. It stops short of stating idempotency or what the call returns.

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?

A single dense paragraph, front-loaded with the core action, then mechanics, then usage rules, then a security warning. Every clause carries information, though the mechanics and policy sentences are packed tightly enough to be a slightly heavy read.

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?

No annotations and no output schema, so the description must cover behavior and results. It explains the delivery path and usage well but leaves 'replyBy' unexplained and doesn't say what the caller gets back (note id, confirmation, or nothing), leaving a real gap for an outbound side-effecting 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 only 25% – only 'request' is documented in the schema. The description never explains 'replyBy', 'title', or how the 'type' enum values map to the described use cases (errand/question/status), so it fails to compensate for the 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?

Uses a specific verb and resource ('Hand a task to the user's personal agent') and describes the delivery mechanism (desk-folder note, git sync, email via doorbell). It does not name any sibling tool explicitly, so differentiation from handoff_inbox_reply or handoff_status relies on the reader inferring it.

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 enumerates when to use it ('errands outside the computer, questions while the user is away, and status lines') and when not to ('never for building or code work'). It also gives a fallback path when ring.rung is false, but doesn't contrast with any named sibling tool.

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?

With zero annotations, the description carries the full burden. It does disclose one non-obvious trait — the status is for the agent to relay, not delivered directly to the user — but says nothing about whether a new status replaces the prior one, where it is stored, or what the call returns.

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?

A single front-loaded sentence with two illustrative examples; nothing redundant and no wasted preamble.

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 2-required-param tool with no annotations, no output schema and no schema descriptions, the definition should cover the undocumented title field and the relationship to handoff_send/handoff_read. It leaves both 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?

Schema description coverage is 0%, so the description must compensate. It effectively explains "line" (the one-line status, with examples) but never mentions "title" at all, leaving a required 120-char field entirely undocumented.

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 ("Post a one-line status") with concrete examples that make the intent unmistakable. It never distinguishes itself from the sibling handoff_send, so an agent cannot tell from the description why it would pick this over a generic send.

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 examples ("Finished phase 1", "Blocked on X, need your call") implicitly sketch the usage context, but there is no explicit when-to-use rule, no exclusions, and no pointer to handoff_send or the other siblings as alternatives.

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?

With no annotations provided, the description carries the burden and does well: it discloses the 15-second poll interval, the default (40) and hard cap (50) on `seconds`, and the rationale for the cap (clients abandon calls near 60 seconds). It also flags that returned text is untrusted third-party content and must not be acted on as instructions, plus notes the shared-repository read path first. Missing only latency implications for the caller.

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 core behavior and polling cadence before the budget constraint and the security caveat. Three sentences and roughly ninety words is proportionate to a blocking, auto-repolling call; no sentence is filler, though the untrusted-content warning is verbose relative to the rest.

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?

No output schema exists, and the description covers the return contract itself: 'Returns the item and whether it was answered.' Combined with the timeout budget and repoll instruction, an agent has everything needed to call this correctly in a loop.

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%, so the description must compensate, and it does for `seconds` — default 40, max 50, min implied by the cap, plus the reason for the cap. The required `id` parameter is only implicitly covered as the target 'desk item', which is inferable but not stated.

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 ('Wait for the agent to answer a desk item') and immediately characterizes the mechanism: re-reads every 15 seconds until no longer waiting. That polling behavior alone distinguishes it from read-once siblings such as handoff_read and handoff_status.

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 context is clear (you are blocked on another agent's answer) and continuation is explicit: 'If `answered` is false, call again to keep waiting.' It does not name the read-once alternative (handoff_read) or state when not to use this tool, so it stops short of full alternative routing.

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