set_pitch

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

With no annotations, the description carries the full burden. It does disclose that this is a destructive "Replace" of an existing curve and adds a real constraint absent from the schema ("At least 2 points"), plus the fact that snap_first overrides the first point's y. It says nothing about what happens on invalid indices, whether the note must already exist, permissions, or what the call returns.

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 purpose is front-loaded, the point-count constraint follows immediately, and the two examples earn their space by showing realistic scoop and fall shapes. The bilingual duplication of the first sentence adds length without adding information for most agents, but nothing is padded or circular.

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 mutation tool with no annotations and no output schema, the definition covers the tricky parts (snap_first, point format, minimum point count) but leaves the identity parameters `part` and `indices` undefined and gives no return or failure behavior. An agent could construct a plausible call from the examples but could not be confident it targets the intended notes.

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 reported as 0%, so the description must compensate. It does this well for points (x/y semantics are demonstrated through the two worked examples, including the shape:"i" usage) and thoroughly for snap_first. However, `part` and `indices` are never explained — an agent is left guessing what indices refer to (note indices? ordering?) and how part scoping interacts with them.

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?

"Replace the pitch curve of notes" is a specific verb+resource, and the parenthetical (scoops, falls, drifts, overshoot) tells an agent exactly which musical objects this covers. It does not explicitly distinguish itself from siblings like set_curve or set_note_expression, so an agent must infer the boundary between pitch-curve editing and parameter/expression editing.

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 snap_first=true/false explanation is genuine when-to-use guidance: true for a normal slide off the previous note, false for a scoop or a hand-set start, each backed by a concrete example. What is missing is any routing guidance versus sibling tools (set_curve, set_note_expression) or a statement of when this tool should not be used.

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