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

Annotations already declare destructive=true and idempotent=true, but the description adds non-obvious behavior: actions execute in order and are NOT atomic, the first refusal stops the rest, 'start' never replaces an existing session, and request_id reuse is only for retries. That is exactly the kind of partial-failure and retry semantics an agent needs before calling a mutating tool.

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 the core purpose, then the invocation rules, then the action catalogue — a sensible ordering with almost no filler. The action list is long but each entry carries distinguishing semantics rather than restating the enum verbatim.

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 multi-action mutation tool with no output schema and 20+ nested action properties, the description covers sequencing, failure, versioning and start semantics well. It stops short of describing error/refusal response shape or how to resume after a mid-batch refusal, which would complete the picture.

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 only 33%, but the description compensates: it explains request_id lifecycle, ties expected_version to read_live_workout's state_version, and disambiguates action types the bare enum cannot (log_set_shorthand = next planned set, start alone or template_id vs next routine workout). Many nested per-action fields (set_id, position, measure, reason) still rely on the schema.

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+resource ('Start a workout or act on the one in progress') and enumerates the 15 action types, so an agent can distinguish it from read_live_workout (read) and log_workout. The scope 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?

Gives concrete operational rules: send a NEW request_id each turn (same one only to retry), pass state_version from read_live_workout as expected_version, and 'start' must be alone in its call. It does not explicitly say when to prefer this over the sibling log_workout, which is the one remaining gap.

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