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

Annotations already declare destructiveHint=true and idempotentHint=true, so safety is covered; the description adds genuinely new behavior beyond them — permanent, irreversible deletion and a filtering rule that silently skips non-gray ids. It does not describe how skipped ids are reported back, keeping it at 4.

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 short paragraphs with zero filler; the destructive nature and the precondition are both front-loaded before the detail.

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 destructive, low-parameter mutation with full schema coverage, complete annotations, and an output schema, the description supplies everything else an agent needs: irreversibility, the approval precondition, and the id-filtering rule.

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, but the description adds real meaning: the ids are validated against gray-underline status and are the exact set deleted. It also clarifies document scope ('the Skim document') beyond the schema's default-to-front-document note.

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 (PERMANENTLY DELETE) and resource (processed/gray underlines) with an explicit scope qualifier (by id) and notes it MODIFIES the Skim document. This cleanly separates it from read-only siblings like get_pending_underlines and the non-destructive mark_processed.

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 a clear precondition ('Only run after the reader has explicitly approved clearing these ids') and a when-not rule ('pending (red) underlines, highlights and other markup are skipped'). It does not explicitly route to a sibling alternative, so it falls just short of a full 5.

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 and closed-world, so the safety profile is covered. The description adds real behavioral context beyond that: the ordering guarantee ('frontmost first') and the exact fields returned including 'current page'.

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?

One sentence, front-loaded with the action and resource, and the field enumeration is a compact literal list. No wasted words.

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?

A zero-parameter read tool with a full output schema and complete annotations needs very little prose; the description covers what is listed, the ordering, and the source application. An agent has everything required to call it correctly.

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?

The tool takes zero parameters, so the baseline is 4. Nothing in the description needs to explain inputs, and the listed fields describe the output rather than a parameter.

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 and resource ('List the PDFs open in Skim') with the result set enumerated, so the agent knows exactly what comes back. It does not differentiate itself from the sibling get_reading_context, which is the one tool an agent could plausibly confuse it 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 implied by the resource description (call this when you need the currently open PDFs), but the description gives no explicit when-to-use, prerequisites, or alternative such as get_reading_context for reading state.

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=true, idempotentHint=true and destructiveHint=false, so the safety profile is covered. The description adds useful return-shape context (reading order, quoted text) but says nothing about pagination, scope limits, or what 'pending' means operationally.

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 naming the resource, ordering, and payload; every clause earns its place 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 a one-optional-parameter read-only tool with a declared output schema and full annotation coverage, the definition supplies enough to invoke it correctly. The only gap is disambiguation from get_reading_context, which the description leaves 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 100%, so document_path and its 'default: the front document' behavior are fully documented in the schema. The description only alludes to 'a document' and adds no syntax or fallback detail beyond what the schema already provides, 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?

The description states a specific verb and resource ('List pending (red) underlines of a document') plus two extra qualifiers: reading order and quoted text. This clearly separates it from the mutating siblings mark_processed and clear_processed, though it never explicitly contrasts itself with get_reading_context.

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: 'pending' suggests items not yet processed, which loosely links to the mark_processed/clear_processed siblings. There is no explicit statement of when to call this versus get_reading_context, nor any prerequisite or exclusion.

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=true and idempotentHint=true, so safety is covered. The description adds real value beyond that: the selection is 'explicitly empty when nothing is selected', the image is 'ground truth for math' while 'the text layer may garble formulas', and the blue/red outline semantics are disclosed.

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 a one-line summary, then a structured enumeration of return contents. Dense but every clause carries information; no filler sentences. Slightly long but justified by the breadth of returned data.

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 full burden of describing return values, and it does so thoroughly: document name/path, page, selection, underlines with quoted text/page/color/comment, text layer, and PNG outlining. An agent knows exactly what to expect from a call.

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 three parameters (page, document_path, include_image) are already documented with defaults, including the fallback logic for page and document_path. The description adds nothing parameter-specific, 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?

States a specific resource (reading context) and enumerates its contents: front document, current page, selection, pending underlines, text layer, and PNG. This implicitly separates it from the narrow siblings like get_pending_underlines and get_open_documents, though it never names them directly.

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 opening phrase 'Everything about what the reader is looking at, in one call' implies this is the comprehensive aggregator, which hints at when to prefer it over the narrower siblings. However, it never explicitly states when to use this vs. get_pending_underlines or get_open_documents, nor any exclusions.

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 declare it is a non-readonly, idempotent, non-destructive write; the description goes further by scoping the mutation (only pending red underlines, other colors skipped, never changed) and explicitly stating nothing is deleted. These details materially reduce the risk of unintended edits beyond what the annotations 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?

Three short sentences, front-loaded with the action and the side-effect warning before the precondition. Every sentence carries distinct information with no padding.

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 rich annotations, a full input schema, and an output schema present, the description supplies exactly what structured fields cannot: the approval prerequisite, mutation scope, and non-deletion guarantee. Nothing needed to invoke it safely is missing.

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 both parameters (ids, document_path) are already documented, including the source of ids and the default front-document behavior. The description's "by id" adds no syntax or format detail beyond the schema, 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?

States a specific verb and resource: recolor pending (red) underlines to gray by id. It is clearly distinguishable from siblings get_pending_underlines (read) and clear_processed (removal), and the title reinforces rather than restates the operation.

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?

"Only run after the reader has approved" gives an explicit precondition for invocation, which is meaningful for a mutation tool. It does not, however, name sibling alternatives or explain the relationship to clear_processed, leaving some routing inference to the agent.

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