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

With no annotations, the description carries the full burden. It does disclose two real traits: results are capped at 8 and each row is dated and attributed, which tells the agent the output is time-scoped and provenance-tagged. It omits permission requirements, what happens when no run exists, and whether the sample is deterministic.

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 scope ('up to 8 sample rows from the last run') front-loaded and the alternative pointer last. Nothing is padded, though the phrasing is clipped enough that some meaning requires inference.

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 2-param, no-output-schema, no-annotation tool, the description covers what is returned and its limits but leaves the required 'source_id' undefined and does not clarify how a source or its 'last run' is identified. Adequate but with 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?

Schema coverage is 50% — 'limit' is documented (default 8, max 8) and the description's 'up to 8 sample rows' corroborates that cap. 'source_id' is undocumented in both schema and description, so the description only partially compensates for the coverage gap.

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 states specifically what is returned: a capped sample (up to 8) of rows from the last run of a named source. That is enough to distinguish it from list_sources and get_source, though it never names a retrieval verb explicitly and the title ('Take the rows from a source') is clumsy.

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?

It gives one routing hint: for more than a sample, use the REST rows endpoint with an instant free key. That is a useful fallback cue, but it does not explain when to prefer this tool over siblings or what preconditions (valid source, prior run) apply.

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