stream_start

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

Annotations already declare the mutation/idempotency profile, and the description adds real behavioral detail beyond them: the buffer is limited and in-memory (results may be truncated or lost), polling is best-effort rather than a tick feed, and the stream terminates on duration expiry or error. It omits how the buffer is later read (stream_poll) and any resource/rate implications, which keeps it below a 5.

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 with the core behavior front-loaded, then the polling caveat, then the termination condition. No filler and every sentence carries new information.

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 5-parameter tool with a nested object, no output schema, and zero schema description coverage, the description covers lifecycle start and stop but not the retrieval path (stream_poll) or what the buffer actually yields to the caller. The stream lifecycle is only half-told, which matters for correct end-to-end use.

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 0% across 5 parameters, so the description carries the burden and only partially does so: it explains tool (must be read-only), interval (bounded), duration (stop condition), and buffer (limited). The nested `arguments` object, which is the pass-through payload for the target tool, and buffer_size limits go entirely unexplained.

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 concrete action (poll a read-only tool at a bounded interval, retain an in-memory buffer) rather than restating the name. An agent can distinguish it from stream_poll/stream_stop/stream_list by the "start a stream" framing, though the description never names those siblings 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?

The use case is implied (repeated polling of a read-only tool) and the caveat "This is browser polling, not a guaranteed tick feed" tells the agent what not to expect. However, there is no explicit guidance on when to choose this over stream_poll, stream_list, or a one-shot data_get_* call, and no stated prerequisites such as the target tool needing read-only semantics beyond the single phrase "read-only tool".

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