graph_build

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

With no annotations provided, the description carries full behavioral disclosure. It clearly states the caching side effect ('cache it'), the report contents ('networks, junctions, total length, and the open ends'), the T-junction vs. break classification logic, tolerance behavior via the 'max-jump rule,' and that 'lengths are exact' in drawing units. This is comprehensive transparency.

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 long but highly information-dense. It is front-loaded with the core purpose, then details behavior and parameters. Every sentence adds meaningful value (e.g., tolerance rule, report details, units). It could be slightly more concise (e.g., splitting into separate sentences for parameters), but it earns its length given the tool's complexity.

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?

The tool is complex (graph analysis with tolerance, break detection, caching), and the description covers all major aspects: what it reports, how tol works, window/layer scoping, exactness of lengths, and the defect-detection purpose. The output schema exists, so detailed return structures need not be repeated. This is complete for an agent to invoke 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?

Schema description coverage is 0%, so the description must explain all parameters. It does: tol's default behavior and override semantics ('tol=None (the default) makes the joining tolerance EMERGE... pass a number only to override'), window's format and scope ('window=[x1,y1,x2,y2] limits it to one drawing area'), and layer's scope ('layer to one layer'). This goes well beyond the schema's bare type definitions.

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 opens with a specific verb and resource: 'Read the 2D geometry as a CONNECTIVITY GRAPH and cache it.' It clearly defines the tool's purpose as converting geometry into a graph structure and analyzing it, distinguishing it from sibling drawing/analysis tools like graph_trace or entity_length by emphasizing the connectivity graph and break detection.

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 provides strong context for when to use the tool: 'so the drawing can be asked questions instead of read' and 'which on a schematic is the defect you were looking for.' It implies a schematic-checking use case. However, it does not explicitly name alternatives or state when not to use it, so it misses the 'exclusions' aspect of a 5.

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