humanize_midi_notes

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

Annotations are minimal (readOnlyHint=false, destructiveHint=false), so the description carries the burden of behavioral disclosure. It adds 'guarded deterministic' and 'complete native note readback', which hint at safety and output, but it does not detail the two-phase commit, token validation, or the fact that applying changes is a mutation. While not contradictory, it leaves important behavioral aspects to the schema.

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?

The description is a single dense sentence that effectively front-loads the core action ('Plan or apply') and embeds key qualifiers (guarded, deterministic). It avoids redundancy and is reasonably concise given the tool's complexity, though it could be clearer with a structured mention of the two modes.

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?

This is a complex tool with 11 parameters, a dryRun/confirmationToken pattern, and no output schema. The description fails to explain the two-phase workflow (call with dryRun=true first, then apply with token), which is essential for correct usage. It also doesn't summarize the validation rules (e.g., maxTimingOffsetBeats constraint) beyond what the schema holds. The high-level flow is missing, leaving agents to infer from the schema alone.

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 100%, so all 11 parameters are individually documented. The description itself does not add extra meaning beyond what the schema provides—it mentions 'clip' and 'grid' which map to clipId and gridBeats, but these are already in the schema. It does not clarify parameter relationships like maxTimingOffsetBeats must not exceed half gridBeats, which is in the schema anyway. Baseline 3 is appropriate.

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 clearly states the tool plans or applies MIDI timing and velocity humanization, with specific mention of clip, grid, and note readback. It distinguishes from generic 'humanize' by including 'guarded deterministic' and the dual plan/apply nature, though it does not explicitly name sibling alternatives like plan_midi_humanization.

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 does not provide explicit guidance on when to choose this tool over siblings such as plan_midi_humanization or apply_midi_velocity_curve. The dryRun/confirmationToken flow is implied but not explained, nor are any scenarios where a separate plan tool would be preferred. Agents may struggle to determine the correct usage context.

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