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

With no annotations, the description carries the full burden and does so richly: it specifies the signed 'fire' payload (run_id, started_at, deadline), the 'clear' payload with late:true and runtime, the superseded-start behavior with no callback, idempotent re-registration returning the same watch, and the billing impact (1c/day, counts toward the free 5).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?

Four dense sentences, front-loaded with the trigger condition and then ordered by lifecycle (register, start, finish, timeout, supersede), so the sequence reads naturally. The billing shorthand and compressed clauses are jargon-heavy but each sentence carries distinct 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?

With no output schema and no annotations, the description covers the callback contract, supersede semantics, idempotency, and cost, which is most of what an agent needs. It does not describe what the registration call itself returns or any error/auth failure behavior, leaving a modest gap for a tool that provisions a persistent watch.

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%, so the description must compensate, and it does: it supplies the 60–86400 range for max_runtime_s (absent from the schema), explains label's per-job scoping and idempotency semantics, and implies callback_url's role via the POST description. The callback_url format and expected response are still left implicit.

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 precise verb and trigger condition ('Fire when something is STILL running past its deadline') and immediately positions it against the sibling concept ('the other side of a deadman'), which separates it from deadman_checkin and content_watch. An agent can identify the tool's role without opening the 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 an explicit workflow: register once, then call overrun_start when a run begins and overrun_finish when it ends, naming the exact sibling tools. It implies the deadman contrast but never states an explicit 'do not use this if...' exclusion, so it stops just short of a 5.

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