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

Annotations already declare readOnly=false, destructive=false and closed-world, so safety is largely covered. The description still adds real behavior beyond them: it discloses that existing beat markers are replaced, and that the density is controlled in musical terms (every beat / 2nd / 4th = a bar in 4/4). The replacement clause sits slightly in tension with destructiveHint=false, but only for derived marker data, so this is disclosed behavior rather than a contradiction.

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 tight sentences, zero filler, and the core capability plus the musical-unit clarification are front-loaded. The overwrite warning is appended where it belongs, at the end.

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 5-parameter mutation with no output schema and 20% schema coverage, the description covers the central concept (beat density) and one destructive side effect, but leaves the start/end range semantics, clipId requirements, and the relationship to get_beats unaddressed. It is enough to understand the tool, not quite enough to call it confidently.

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 20% — only projectId is documented — so the description must compensate, and it does for the hardest parameter: "every" is explained in musical terms (1 = every beat, 4 = once a bar in 4/4), which the bare integer 1-16 does not convey. start/end and clipId remain unexplained, leaving a real gap for a time-range tool.

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 ("Put markers on the beats of a music clip") plus the rationale ("so cuts can snap to them"), which is far more specific than the sibling add_marker/add_note family. It never names a sibling to contrast with, so an agent must infer the difference between this and add_marker (single marker) or get_beats on its own.

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?

Usage is implied rather than stated: the tool exists so cuts can snap to beat positions, but the description gives no conditions for choosing it over add_marker, get_beats, or batch_edit, and no prerequisites (e.g., whether beat detection must run first). An agent can guess the context but is given no exclusions.

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