Kinro

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

Annotations only give non-read-only, open-world, non-idempotent, non-destructive; the description adds far more: carriers receive the business's details, carriers may not contact the user, Kinro contacts the user off-chat only at agreed contact details, E&O insurance backs the advice, calls are never retried, and the test endpoint still creates a real request. This is exactly the extra context annotations cannot carry.

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?

Long, but front-loaded with purpose and routing before the operational caveats, and most sentences carry genuinely non-obvious behavior. There is some redundancy around follow-up questions and carrier independence that could be tightened.

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 an open-world, asynchronous, consequential tool this covers everything an agent needs: conversation lifecycle, retry prohibition, deferred answers and the retrieval path, privacy constraints, test-endpoint conduct, and how to handle returned quote links. An output schema exists, so return-shape explanation is unnecessary.

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 message, files, and conversation_id in depth. The description adds workflow context (first call opens the request and saves what the user shares) but no additional field-level syntax or constraints beyond what the schema states.

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+resource: sending a message to Kinro, a licensed business insurance broker, to get coverage advice, carrier eligibility answers, and quotes. It clearly distinguishes itself from the sibling get_kinro_messages, which it describes as the retrieval path when Kinro is still working.

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?

Explicit routing: 'when Kinro says it is still working, call get_kinro_messages with wait set to true'; 'Each call sends a new message, so do not repeat a call to retry it'; 'The first call opens a Kinro insurance request'. It also gives conditions for the test endpoint, how to handle unknown facts, and what to do with quote links.

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: default limit of 5, up to 100, new messages always included, wait returns as soon as a message arrives or after a short wait with a note, and it sends nothing to Kinro. Annotations declare readOnlyHint=false and idempotentHint=false, which the description does not explain; the 'sends nothing' clause partially offsets that, but doesn't address non-idempotence.

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-loads what the tool returns, then covers limit, wait, and recovery use cases. It is slightly dense and repeats the limit range already in the schema, but every sentence conveys usable behavior.

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?

Covers the return contents, ordering, new-message marking, default and maximum limit, wait semantics, and recovery use case. With an output schema present the return shape needn't be explained, but the description does not reconcile readOnlyHint=false and idempotentHint=false with its 'read' framing, leaving a small gap.

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 schema already documents wait, limit, and conversation_id in detail. The description reinforces limit (default 5, up to 100) and wait behavior but adds no syntax or format beyond the schema, 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 (read) and resource (this conversation with Kinro), specifies what is returned (the user's messages and Kinro's, oldest first, new ones marked). It is clearly distinct from the sibling ask_kinro, which the description implicitly contrasts by noting 'It sends nothing to Kinro.'

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?

Explicit when-to-use guidance: set wait=true when Kinro says it is still working, and use it to recover a reply or quote you missed or lost. This names the concrete conditions that select the tool and even clarifies its non-sending nature relative to ask_kinro.

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