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

With no annotations, the description carries the full burden and does so: it discloses the connection rule ('endpoint 0 to the previous endpoint 1', 'All curves turn left'), the validation thresholds (0.1 mm, 0.1 degrees, crossings, overlaps), the failure contract ('returns measured errors and writes nothing'), and the success side effect (creates a unique .xtc under layouts/). This is exactly the mutation/side-effect context an agent needs before calling.

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?

Front-loaded with the core action, then geometry rule, example, constraints, and outcome contract in that order. Every sentence adds information an agent would otherwise have to guess; the example is compact and directly illustrative rather than padding.

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-value detail is optional, yet the description still summarizes path and parts list. Validation rules and the write-nothing-on-failure guarantee are fully covered. The remaining gap is the unexplained board-size parameters, which matter because they interact with the centring margin and overlap checks.

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%, so the description must compensate. It adds real meaning for 'parts' (ordered, catalogue part numbers, ordering drives endpoint chaining), but width_mm and height_mm are never named or explained beyond an oblique 'centres the result on the board with a fixed 50 mm centreline edge margin' — their units, defaults (2000x1200) and effect on validation are left to the schema.

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 ('Build a flat closed loop from an ordered list of catalogue part numbers') and immediately distinguishes itself from the sibling list_parts, which only enumerates parts. An agent knows exactly what artifact is produced (.xtc layout) without opening the schema.

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?

Gives a concrete worked example ('six 24130, one 24188...') plus explicit exclusions ('Does not open XTrackCAD or check rolling-stock clearance'), which tells the agent when this tool is the wrong choice. It does not name an alternative tool or state preconditions such as requiring parts to exist in the catalogue, so it stops short of full routing guidance.

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

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

No annotations exist, so the description carries the full burden and does a solid job: it declares units (mm), the endpoint 0-in / endpoint 1-out sequence convention, and that curves turn left. It stops short of stating that this is a pure read with no side effects or how results are ordered/limited.

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?

Purpose is front-loaded in the first sentence, and every subsequent sentence adds a distinct constraint (units, endpoint convention, unsupported categories). The telegraphic style is efficient, though the geometry sentence is only loosely tied to a lookup tool's primary job.

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?

An output schema exists, so return values need not be explained, and the description covers the domain constraints an agent would otherwise get wrong. What remains thin is query-matching semantics for the one undocumented parameter, but overall it is sufficient to call the tool correctly.

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?

The single 'query' parameter has 0% schema description coverage, so the description must compensate. It partially does by saying the query can be a part number or a description, but it gives no guidance on matching behavior (exact vs. substring), case sensitivity, or what an empty default query returns.

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?

Specific verb ('Find') plus a tightly scoped resource ('supported ordinary Märklin C-track straights/curves') and the lookup key ('by number or description'). The scoping language also implicitly separates it from the sibling build_layout, which assembles rather than looks up parts.

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?

Explicit exclusions are stated: turnouts, special tracks and catalogue editing are unsupported, which is real when-not guidance. It does not explicitly say when to reach for build_layout instead, so the routing to the sibling is left for the agent to infer.

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