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

Beyond the annotations it discloses cost ("counts as one song"), latency ("about a minute"), UI feedback ("the player shows it"), the exact blast radius (other four stems unchanged) and reversibility via undo_stem. That reconciles destructiveHint=true with a safe, undoable operation — exactly the context annotations cannot convey.

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 action and scope, then the cost/latency/undo facts, each earning its place. The semicolon-chained phrasing makes "the mix becomes the new part with the rest" harder to parse than needed.

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 a simple 3-param schema, annotations and no output schema, the description covers what an agent needs: side effects, credits, wait time, and recovery. It leaves the version parameter and any failure conditions (unfinished song, invalid stem) unaddressed.

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 100% and both enums are self-documented, so the baseline is 3. The description's "the other four" implicitly confirms the five-part enum, but the version parameter and its interplay with redo are never mentioned, so it adds no meaning beyond 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?

"Make one stem of a finished song again" gives a specific verb, resource and scope (one stem, remaining four untouched). It names the exact alternative (undo_stem) and the inverse outcome, so an agent can distinguish this from every sibling without opening a 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?

It establishes the operating context clearly — the song must be finished, only one part is regenerated — and explicitly routes the reverse action to undo_stem. It stops short of stating when not to use it (e.g. still-generating songs) or whether repeated redos are allowed.

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