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

It adds useful policy information ('Always allowed') and clarifies a key behavioral quirk: how the `in_reply_to` parameter handles threading and defaults to the original sender, which is behavioral context beyond the schema.

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 highly concise, utilizing only two sentences to capture the core functionality, the policy, and the complexity of reply threading. Every word contributes to clarity.

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?

Given the rich schema and annotations, the description completes the picture by explaining the operational context of drafting, especially regarding how reply threads are established and the default behavior of the 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?

Although schema coverage is 100%, the description adds semantic value by explaining formatting (Markdown vs HTML rendering) and crucial usage details for parameters like `to` (recipient format and empty list default) and `subject` (automatic prefix when replying).

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 clearly states the primary function as saving a message to the Drafts folder. It differentiates this action (draft creation) from immediate actions like sending, implicitly setting it apart from the send sibling.

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 defines the use case (saving for later review) which is the primary differentiator from sending, but it does not explicitly state when *not* to use it or compare its usage requirements against alternatives like `send`.

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 openWorldHint=true, so the safety profile is covered. The description adds useful returned-data context (special-use labels, message/unread counts), but it doesn't say whether folders are returned flat or nested, or whether system folders are included by default.

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 no filler; the resource comes first and the return content follows. Every clause carries information an agent cannot get from the schema or annotations.

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 zero-parameter, read-only enumeration with no output schema, the description covers what the tool returns well enough to invoke confidently. The main remaining gap is folder hierarchy and whether ordering or nesting is guaranteed.

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, which is the baseline-4 case; there is no parameter semantics to explain. The description correctly signals that the call is unconditional and returns the entire folder set rather than a filtered subset.

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 (List) and resource (mail folders) and even previews the returned fields: special-use flags plus message and unread counts. Nothing among the siblings (draft, read_attachment, read_message, search, send) overlaps with folder enumeration, so an agent can distinguish it, though the description never states that differentiation explicitly.

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?

There is no explicit statement of when to call this versus alternatives, but usage is strongly implied: an agent that needs folder identifiers before searching or reading messages would call this first. 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?

Annotations already declare readOnlyHint and openWorldHint, so the lower bar applies; the description still adds real value by disclosing that only certain MIME types yield text and that others return metadata only — a behavioral outcome the agent cannot infer from annotations. It omits that output is truncated at max_chars (default 20k), though the schema carries that.

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, zero waste, with the core action and the format caveat front-loaded before the short provenance note. The parenthetical format list earns its space by defining the tool's actual capability boundary.

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 does cover the key return-shape distinction (text vs. metadata-only), which is the main thing an agent needs to interpret results. It could have mentioned that long extractions are truncated per max_chars, but overall it is sufficient for correctly invoking and reading this 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%, so the baseline is 3. The description's note that the index originates from read_message duplicates the schema's own wording, and it says nothing about uid, folder, or max_chars, so it adds no meaning beyond the structured field definitions.

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 ("Read the text of one attachment of a message") and immediately bounds the scope with the supported format list plus the fallback behavior for other types. An agent can distinguish this from read_message (message body) and from list_folders/search without opening a 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?

"Index comes from read_message" supplies the prerequisite workflow, making clear this is a follow-up call to a message read rather than an entry point. It stops short of naming exclusions (e.g., 'do not use for message bodies, use read_message'), so it is clear context but not a full when/when-not routing statement.

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?

The annotations declare readOnlyHint=true, but the description and schema state that mark_read=true sets the message's read flag, which is a state mutation. This directly contradicts the read-only annotation, even though the default behavior leaves the message unread.

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 a single well-structured sentence with the core action and returned data front-loaded. Every clause adds useful information without excess.

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 low-complexity read tool, the description adequately explains what is returned and the default mark_read behavior. It omits some edge details such as truncation and error cases, but those are partly covered by the schema.

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 parameter meanings are already documented. The description adds some context about mark_read's default effect and the returned body/attachment structure, but it does not add parameter-level detail beyond what the schema provides.

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: read one message, and it enumerates the returned content: headers, converted body, and attachment list. It is clear but does not explicitly distinguish itself from sibling tools such as read_attachment or search.

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 does not say when to use this tool versus search, read_attachment, or other siblings. The schema mentions that uid comes from search, but the description itself provides no explicit usage context or 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 already declare readOnlyHint=true and openWorldHint=true, so the safety profile is covered. The description adds genuinely useful behavior beyond that: the exact return fields, newest-first ordering, and the INBOX default — valuable because there is no output schema. It stops short of describing pagination behavior or what happens when results exceed the limit.

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 with zero waste: the operation and its filter axes come first, then the return shape and ordering. Nothing is repeated or 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?

For a 9-parameter, zero-required search tool with no output schema, the description covers the important context: default folder, date format, return fields, and sort order. The one remaining gap is how to page past the 100-message ceiling, which an agent would need to know for broad queries.

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 every parameter is already documented with its semantics and defaults. The description restates the filter set and date format but adds no syntax or edge-case detail beyond the schema, so 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?

States a specific verb (search) and resource (a folder of mail) plus the exact filter axes and returned fields. It implicitly separates itself from read_message by returning only headers/metadata (uid, date, from, subject, size, has_attachments) rather than bodies, but it never names the sibling it is not, so differentiation is inferred rather than stated.

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: search mail in a folder, defaulting to INBOX, with since/before in YYYY-MM-DD. There is no explicit when-to-use guidance versus read_message, list_folders, or read_attachment, and no guidance on how to retrieve more than the 100-result cap.

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 meaningful behavior beyond annotations: the message is copied to Sent, sending is conditional on server permissions and allowed recipients, and draft is the fallback. Annotations already cover the destructive/non-idempotent/open-world profile, so the description need not repeat those traits.

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 compact sentences with the main action and constraints front-loaded. Every clause contributes useful routing or behavioral information without 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 destructive send operation with rich annotations and fully documented inputs, the description covers the key operational constraints, side effect, and alternative tool. No output schema exists, and the agent has enough context 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?

Schema description coverage is 100%, so the schema already documents all six parameters, including reply metadata, recipients, and body formatting. The description adds no additional parameter semantics beyond what the schema provides, so baseline 3 is appropriate.

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 action and resource: sending a message over SMTP and copying it to Sent. It also distinguishes itself from the sibling draft tool by naming draft as the fallback when sending is not allowed.

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 gives the conditions for use: the server must allow sending and recipients must be allowed. It also names the alternative tool, draft, for the otherwise case, leaving little inference needed.

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