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

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so the safety profile is covered by structured data. The description adds the genuinely non-structured behavior of polling cadence and stopping on terminal states, but one clause points at poll_after_seconds, which appears nowhere in the schema and reads as dangling.

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?

Two short, front-loaded sentences with no filler, and the non-mutating claim comes first. But the second sentence is cryptic rather than dense: it references an undefined field and an unspecified notion of "terminal states," so the brevity costs clarity.

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?

An output schema exists, so return values need not be explained, and the tool is a simple two-param read. Still, with zero parameter descriptions in the schema, the definition never explains what project_id and task_id identify, nor which states count as terminal.

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?

Both required parameters (project_id, task_id) have 0% schema description coverage, so the description carries the full burden and supplies no meaning at all for either. Instead, it names poll_after_seconds, which is not a parameter, creating confusion about the actual inputs.

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?

"Read saved progress" names a specific verb and resource, and "without modifying state" pins down that it is a non-mutating status read. It does not differentiate from siblings such as research_list_tasks or research_get_report, so an agent must infer the boundary, but the core operation is unambiguous.

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?

"Poll no faster than poll_after_seconds; stop on terminal states" gives real polling and termination guidance, which is the key usage advice for a status tool. However, it offers no when-to-use versus the many siblings (research_list_tasks, research_cancel_task, research_get_report), and the referenced poll_after_seconds is not a parameter of this tool, leaving the cadence source unexplained.

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