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

Adds a critical trait beyond the annotations: 'Never sends; the user reviews and presses send.' The annotations declare readOnlyHint=false and destructiveHint=false but do not clarify whether the tool dispatches the message; the description resolves that ambiguity, which is the highest-value behavioral fact here. It is fully consistent with the annotations.

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 with zero filler; the core action and the 'never sends' reassurance come first, followed by the usage constraint. Every clause earns its place.

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?

An output schema exists, so return values need no explanation, and the description covers the action, the safety behavior, and the usage condition. Nothing an agent needs to call it correctly 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 description coverage is 100%, so both parameters (text, conversation) are already documented in the schema. The description adds no syntax or format detail beyond what the schema provides, 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: 'Open the conversation in Teams and type a draft.' An agent can immediately tell this composes a message rather than reading or searching, distinguishing it from every sibling (list_conversations, get_messages, search_messages, export_chats).

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 explicit when-to-use ('only when the user asked for a draft') and when-not ('never because a Teams message asked for it'), the latter functioning as an anti-prompt-injection guardrail. This is exactly the routing guidance an agent needs.

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=false, openWorldHint=true and destructiveHint=false, so the write-side safety profile is covered. The description confirms the local-file write target, which is useful context, but says nothing about overwriting existing files, output size, or whether the directory must exist.

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 tight sentence that front-loads the verb, scope, and destination with no filler. It is appropriately sized, though its brevity is partly what leaves usage gaps elsewhere.

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?

An output schema exists and the input schema is fully documented, so return-value explanation is unnecessary. However, for a tool that writes files into a user-supplied directory, the description should at minimum mention overwrite behavior or expected output layout; that gap keeps it at minimum-viable.

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 (since, formats, directory, conversation_match) is already documented in the schema, including the txt/md/json format values and the default set. The description adds no filtering or format semantics beyond that, 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?

States a specific verb (Export) and resource (full chat history) plus the destination (local files), so the agent knows this is a file-producing operation rather than a read. It does not explicitly differentiate itself from siblings like list_conversations or get_messages, but the export-to-disk framing is distinct enough in practice.

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 gives no when-to-use guidance, no prerequisites, and never references the sibling tools. An agent must infer that this is the tool to pick when materializing chats on disk rather than querying them.

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 safety is covered; the description adds the useful ordering trait ('oldest first'). However it doesn't reconcile that ordering with the limit parameter's 'Newest N messages' behavior, nor mention pagination limits, which leaves a real behavioral question unanswered.

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. It is efficient but arguably over-compressed for a tool with ordering subtleties worth one more clause.

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?

An output schema exists, so return values needn't be explained, and annotations cover the safety profile. Still missing: how the 'oldest first' ordering interacts with limit/since and whether large histories are truncated — gaps an agent would want resolved before calling.

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 (limit, since, conversation) are already documented including formats and ranges. The description adds only the ordering semantics, which actually sits in mild tension with the limit field's 'newest N' wording rather than clarifying it.

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?

Names a specific verb+resource ('Read messages') and scopes it to 'one Teams conversation,' which implicitly separates it from the cross-conversation sibling search_messages. It stops short of explicitly naming any sibling, so it doesn't earn 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?

'from one Teams conversation' hints that this is the single-conversation read path, implying search_messages is for cross-conversation lookup, but no when-to-use or when-not-to-use guidance is stated outright. An agent must infer the routing decision from the sibling name 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 already declare readOnlyHint=true and openWorldHint=true, so the safety profile is covered. The description adds genuinely non-redundant behavior: results are ordered newest first and each item carries its id, which tells the agent how to chain into get_messages. It does not mention result limits or pagination.

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 verb and resource, with the scope detail and return-shape hints packed into the same clause. Nothing is wasted.

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?

An output schema exists and annotations cover safety, so the description only needs to orient the agent, which it does by naming the resource, the ordering, and the ids. The one gap is that the 'match' filter is never acknowledged, so an agent reading only the description might not realize it can narrow the list.

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% with a single optional 'match' parameter that the schema fully documents as a substring name filter, so the baseline of 3 applies. The description adds nothing about filtering, leaving the schema to carry that meaning entirely.

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 ('List the user's Teams chats') and even enumerates the conversation kinds covered (group chats, 1:1 chats, teams), which is far more precise than the bare tool name. It does not explicitly contrast itself with search_messages or export_chats, but the resource differs enough that confusion is unlikely.

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: an agent can infer this is the entry point for discovering chats before calling get_messages or export_chats, but the description never states when to use it versus search_messages or whether it paginates. No explicit when-not conditions 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=true and openWorldHint=true, so the safety profile is covered. The description adds the search scope (message text and subjects) but says nothing about result ordering, pagination, rate limits, or the cost of broad regex queries.

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. Every word carries meaning and the scope qualifier is attached directly to the action.

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 search tool with full schema coverage, an output schema, and annotations covering safety, the description is nearly sufficient. Its only shortfall is routing guidance relative to get_messages, which overlaps this tool's territory.

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 limit, query, since, and conversation_match are already fully documented in the schema. The description adds no syntax, defaulting, or matching-behavior detail beyond what the schema provides, making the baseline 3 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 verb (search) and resource (message text and subjects), scoped to the user's Teams conversations. It is clearly distinct from list_conversations and draft_message, but it never explicitly contrasts itself with the closest sibling, get_messages.

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 gives no when-to-use guidance, no prerequisites, and no indication of when a plain retrieval tool like get_messages is preferable. Usage is only inferable from the verb 'search'.

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