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

Annotations already mark this non-readonly/non-destructive, and the description adds genuinely new behavioral context beyond them: it acts only on the explicit project_id and returns the project identity, it caps flashes for safety, and it adds neither sound nor zoom. It does not disclose reversibility or undo semantics, keeping it below 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?

It is front-loaded with the project-scoping constraint, which is good, but the rest is a dense run-on wall of jargon-heavy clauses with inconsistent structure. Most clauses carry information, but readability suffers and some (e.g. re-listing kind values already in the enum) partially duplicate structured data.

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 16-param, nested-object, no-output-schema mutation, the description covers scoping, placement, density, and safety but leaves more than half the parameters undocumented. An agent could place accents correctly but could not confidently set intensity, count, or seed without guessing.

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 13% across 16 params, so the description must compensate, and it does explain at (program seconds), on (beat|downbeat|onset restricted to user-supplied tracks), density (sparse = strongest, >=2s gaps), scope='all', and region {shape:'card'}. However many params — intensity, count, seed, bands, colour, frames, decay_f, start/end — get no explanation anywhere, leaving substantial gaps.

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 states a specific verb+resource ('1-N-frame punctuation on the speech') and enumerates concrete accent kinds, and it distinguishes itself from siblings by declaring 'Adds no sound or zoom' (vs add_sfx/add_zoom/add_strobe). The heavy jargon makes the core action slightly opaque on first read, but the resource and effect family are identifiable.

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 gives actionable when/when-not rules: 'One punctuation family per section; never inside a payoff hold', 'beat|downbeat|onset only on a track the USER supplied', and 'Flash safety keeps <=3 full-frame flashes a second unless the user asked for strobing'. It does not name sibling alternatives to route between, so it stops 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.