llm_job_watch

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint false. The description adds behavioral context beyond those annotations: notifications are ephemeral and only delivered while the watch request remains active, and the progress is privacy-safe and normalized. No contradiction with 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?

Two sentences carry the core purpose and an important behavioral condition without excessive length. The phrase 'privacy-safe normalized progress' is somewhat jargon-heavy but the overall structure is tight and front-loaded.

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 basic invocation with just jobId, the description is probably sufficient. However, with no output schema and unclear semantics for progressLimit and afterProgressSeq, the definition is incomplete for nuanced use and does not explain what notifications or return values actually contain.

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 description coverage is only 25%, with only jobId documented. The description does not explain waitMs, progressLimit, or afterProgressSeq, and does not describe how they influence the watching behavior. With low schema coverage, the description needed to compensate but largely did not.

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 names a specific action ('Wait briefly') and a specific resource ('progress on an owned async job'), and adds the key distinction of notifications being emitted only while the watch request is active. It is clear enough to separate from llm_job_status and llm_job_result, though it does not explicitly name those alternatives.

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?

The description implies when to use it: when the MCP request carries a progress token and the caller wants live progress notifications. However, it does not explicitly say when to prefer this over llm_job_status or llm_job_result, nor when not to use it.

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