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

With no annotations, the description carries the full burden and does well on the most important trait: it never sends, and a human must review and send. It also explains the threading effect of reply_to_uid. It omits return behavior (e.g., draft id), permission/auth requirements, and whether duplicate drafts are created on repeat calls.

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 its most important constraint (never sends), with no filler or redundancy. Every sentence contributes information an agent needs.

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 five-parameter mutation tool with no annotations and no output schema, the description covers the critical behavior (no send, human review, threading) but leaves gaps: it doesn't describe what is returned after creating the draft, how missing cc is handled, or any permission requirements. Adequate but not fully self-sufficient.

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 80%, so the schema already documents to, body, subject, and reply_to_uid, with only cc undocumented in both places. The description's note that reply_to_uid turns the draft into a reply in that message's thread adds real meaning beyond the schema's bare "uid of the message this replies to," but it is a small increment over an already well-covered schema.

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 ("Create a draft in Gmail") and immediately clarifies the crucial constraint that it never sends. It does not explicitly distinguish itself from sibling tools like gmail_search or gmail_read, but the operation is unambiguous enough that an agent won't confuse it with the read-only siblings.

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: "A person reviews and sends it" suggests this is the right tool when human review is desired rather than immediate delivery. There is no explicit when-to-use/when-not-to-use guidance and no mention of alternatives, so the agent must infer the selection criteria.

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?

No annotations are supplied, so the description carries the full behavioral burden. It implies a read-only operation via 'List', but says nothing about whether system labels, nested labels, message counts, or label IDs are returned, nor about result size or pagination for large mailboxes.

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 short sentence with no filler and the resource front-loaded. It is efficiently written, though the terseness borders on under-specification rather than maximal conciseness.

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 tool is simple (zero params, no nested objects), which lowers the bar, but with no annotations and no output schema the description could reasonably state the shape of what comes back (label names/IDs) and any auth requirement. It is adequate but leaves real gaps.

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 schema has nothing to document and the baseline is 4. There is no parameter meaning the description could add or omit.

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 uses a specific verb ('List') and resource ('the mailbox's folders and labels'), so an agent immediately knows this is a read-only enumeration of label/folder names. It stops short of a 5 because it does not explicitly distinguish itself from siblings like gmail_search, though the resource boundary is clear enough to infer.

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 when-to-use guidance, no mention of prerequisites such as authentication or scopes, and no named alternative. An agent must infer that this is the discovery step used before gmail_search can filter by label; that inference is plausible but not stated anywhere in the description.

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 it does disclose one important side effect: the message is not marked as read. It also adds a prompt-injection caution about untrusted email content, which is genuinely useful behavioral context. It omits other operational traits an agent would want: auth/permission expectations, what happens on an invalid or stale uid, and whether large bodies or attachments are truncated.

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 what is returned, then the side-effect clarification, then the safety caveat. No filler and nothing restated from the title or schema.

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-parameter read tool with no output schema, describing the returned fields (headers, text-converted body, attachment list) covers the main gap an agent would otherwise face. The remaining omission — error behavior on a bad uid — is minor given the tool's simplicity.

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 uid parameter is already documented in the schema as coming from gmail_search, so the description only reinforces provenance rather than adding new meaning. Baseline 3 is appropriate when the schema fully documents the parameters.

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 (read one message) with an explicit scope (single message, not a list) and enumerates exactly what comes back: headers, body with HTML converted to text, and attachments. That enumeration alone lets an agent distinguish it from gmail_search without opening either 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 required prerequisite clearly — the uid must come from gmail_search — which effectively documents the search-then-read workflow and points at the sibling tool. It stops short of any explicit exclusion (e.g., don't loop this over every search hit; use it only when the body is actually needed), so it is clear context rather than full when/when-not 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 full burden and largely meets it: it discloses result ordering ('Newest first') and the exact returned fields (uid, date, sender, subject, labels), which usefully implies bodies are not returned. It also adds an untrusted-content warning. It omits pagination behavior, permission scope, and rate limits, so it is strong but not complete.

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, front-loaded sentences: capability and syntax first, then ordering and return fields, then the safety note. Every sentence carries distinct information with no redundancy.

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 simple two-parameter read tool with no output schema, the description supplies the return-field list and ordering that would otherwise be missing. The remaining gap is the max_results cap/default and any pagination behavior, which is minor for a search 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 only 50%, and the description compensates well for the query parameter by supplying real Gmail search syntax examples beyond the schema's 'Query as typed into Gmail's search box'. The max_results parameter receives no explanation of its 1-50 cap or default, leaving one parameter undocumented in both places.

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 ('Search') and a precise resource and scope ('the whole Gmail mailbox'), which implicitly separates it from gmail_read (single message) and gmail_draft. It stops short of naming any sibling explicitly, so the 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?

Two concrete query examples ('from:billing@example.com newer_than:7d', 'subject:contract has:attachment') establish the intended use case, and 'whole mailbox' signals broad retrieval. There is no when-not guidance or explicit routing to gmail_read/gmail_draft, which keeps it below a 5.

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