geometry.matrix_multiply

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

Annotations already cover the safety profile (readOnly, idempotent, closed-world), so the description earns credit for the constraints they don't cover: Decimal-only arithmetic with no float intermediate, hard 200×200 dimension limits, and the requirement that A's column count match B's row count. It omits what happens on a shape mismatch (error vs. silent failure), which is the one behavioral gap left.

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?

Four sentences, front-loaded with the operation, then shapes, then limits, then precision and ordering. Each sentence contributes; the non-commutativity sentence is the only one bordering on obvious, though it does serve as an operand-order warning.

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?

Output schema exists, so return values need no prose, and the description fully specifies input types, shapes, size caps, and precision behavior. The only missing piece an agent would want is the failure mode for mismatched or oversized inputs.

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 0% — the schema only lists 'A' and 'B' as string arrays — so the description carries the full burden and does: it specifies the m×k and k×n shapes, that entries are decimal strings, and the compatibility rule between them. It stops short of explaining why entries are strings or what an invalid entry produces.

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 (matrix product A @ B) computed in Decimal, and immediately contrasts it with adjacent operations by naming operand order and shape semantics. Reading only the description, an agent knows this is a matrix product, not a determinant, inverse, or linear solve, so the sibling tools metaphorically just sit in a different category.

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 fixes operand ordering ('A @ B 와 B @ A 는 다르다') and the shape compatibility rule, so the call context is implied. But it never says when to prefer this tool over matrix_solve, matrix_inverse, or scalar core.mul, and gives no exclusions or prerequisites. Usage is therefore inferred rather than guided.

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