Visa konversation

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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds substantial operational context: nothing is modified and no mail is marked read, searches take 3–8 s on a ~30k-message account, hard caps of ten searches / 200 Message-IDs / 60 s, and that other tools may run between searches. It also defines exactly when complete becomes False and why note explains it.

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?

Purpose is front-loaded and every paragraph is load-bearing, including the prompt-injection warning about subject/display names and the '[U+XXXX]' address encoding. It is nonetheless dense, with security policy interleaved between mechanics and limits, which slightly dilutes scannability for a three-parameter tool.

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?

There is no output schema, so the description supplies the full return shape (mailbox, messages with uid/from/subject/date/message_id/flags, total, complete, note) as well as the ValueError conditions. Combined with the sibling routing and cap behavior, an agent has everything needed to call it and interpret partial results.

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?

With 0% schema description coverage the description must carry the load, and it does: uid is scoped to mailbox, mailbox is the folder where the UID applies, and max_messages gains truncation semantics (start mail plus most recent when the thread is longer). The 1–200 range and default 50 largely duplicate what the schema already encodes, so this falls just short of full marks.

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 ('Hämta hela konversationen (tråden) som ett mejl ingår i') and immediately distinguishes itself from the sibling it could be confused with by noting the body is not fetched and that each message must be read with get_email(uid, mailbox=...). An agent can tell this apart from get_email and search_emails 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?

Explicitly routes the agent: use get_thread for the chain, then get_email per message with the mailbox field from the response. It also states the when-not conditions – messages replying without In-Reply-To/References are excluded, and a start mail lacking a Message-ID means no thread search is possible and only the start mail is returned.

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