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

Annotations already cover readOnly/idempotent/non-destructive, and the description adds genuinely new behavior: the runs:read scope requirement, that analyses run in the background, the polling contract, and the null-state semantics. The only gap is that it does not say what package price/refund fields look like when work is still in progress, but for a read-only status tool this is a minor omission.

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-loaded with what is returned, then polling rules, then boundaries against sibling tools; nearly every sentence carries operational weight. Minor redundancy between 'until it is null' and the closing 'Do not keep polling when null', and the parenthetical field list is dense, but nothing is wasteful enough to cut.

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?

With no output schema, the description takes on the burden of describing return values and does so explicitly, including the meaning of the sentinel field next_poll_after_seconds and the nested package keys. For a one-parameter read tool it covers inputs, outputs, gating semantics, and prerequisites.

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% and the single run_id parameter's origin is already documented in the schema ('from qualitative_list_runs or qualitative_start_analysis'). The description adds no further meaning about run_id format, validity, or error behavior for unknown ids, so the baseline of 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?

Names a specific verb and resource ('Returns the current state of one analysis') and enumerates the exact fields returned (status, package, steps, progress, deliverables, web_url, action_required, next_poll_after_seconds). It also explicitly distinguishes itself from qualitative_list_runs (single analysis) and qualitative_list_deliverables (outputs), so an agent can route correctly without opening a 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?

Gives concrete operational guidance: poll no sooner than next_poll_after_seconds, stop when it is null, and explains the two null causes with the correct follow-up (qualitative_retry_analysis/qualitative_cancel_analysis for retry_or_cancel, tell the user and hand over web_url for continue_in_web_app). It also routes output retrieval to qualitative_list_deliverables, leaving little to inference.

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