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

Annotations already cover the safety profile (readOnly, non-destructive, closed-world), and the description adds real behavior: the 3-run cap, the meaning of `wedged` (no progress for 15 minutes, released at the next start_* call), the null case for `runs_left_today`, and the data-policy boundary ('No market data, only the person's own runs'). These are traits an agent cannot infer from the annotations.

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?

The trigger condition is front-loaded and the prose is readable, but the return shape is stated twice: once narratively ('Returns one row per run: its task_id, its kind, the seconds since it last reported progress, and `wedged`') and again verbatim as a structure literal. That duplication is redundant for a tool with no inputs.

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 input parameters and no output schema, the description carries the full burden and meets it: it documents the return object's shape, the meaning of its non-obvious fields, the limit, and the data policy. An agent can call this and interpret the result without guessing.

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?

Zero input parameters, so there is nothing to disambiguate and the baseline is 4. The description instead spends its effort on return-field semantics (task_id, kind, last_progress_age_s, wedged), which is useful but falls outside parameter semantics proper.

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 and resource ('List the person's runs that are still counted as running') plus a scope bound ('at most 3 at once'), which cleanly separates it from the get_*_progress siblings that inspect a single completed/ongoing resource and from start_* tools that launch runs.

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 an explicit trigger condition: use when a start_* call is refused because 3 runs are already going and it's unclear which, 'most often after a conversation moved on in the middle of one.' It also tells the agent to read it before proposing a run's settings, which is a concrete pre-condition rather than vague guidance.

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