Skip to main content
Glama
kompas-mcp

kompas-mcp

by biryak

Select geometry / reference data

select
Read-onlyIdempotent

Find faces, edges, materials, threads, or fits by geometric properties or KOMPAS reference queries; validate expected counts to catch mismatched candidates.

Instructions

Find faces/edges by geometric properties (never by index), materials already taken from KOMPAS's handbook, or REFERENCE DATA with its source (thread, fit) — quote such values from here, never from memory. Face/edge results are 'topo:' handles valid only for the returned model revision. With expect_count, a mismatch returns SELECTION_MISMATCH listing all candidates with their attributes.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "select"
  }
}

Input Schema

NameRequiredDescriptionDefault
docNoDocument handle (only for face/edge)
kindNoface/edge: geometry; material: materials taken from KOMPAS's handbook; thread: a thread from KOMPAS's thread table (query 'M20x1.5'); fit: limit deviations of a size from KOMPAS's tolerance table (query '50H7', '30h6')face
queryNomaterial search text / thread / size+field
normalNoUnit outward normal, e.g. [0,0,1]
extremeNo
radius_mmNo
created_byNoobj handle of the feature that made it
curve_typeNo
expect_countNoFail with candidates if the match count differs (strongly recommended)
surface_typeNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv0.6.0
    • changedInput schema / properties / doc / description
      Previous value: -"Document handle (not needed for kind='material')"New value: +"Document handle (only for face/edge)"
    • changedInput schema / properties / kind / description
      Previous value: -"What to find"New value: +"face/edge: geometry; material: materials taken from KOMPAS's handbook; thread: a thread from KOMPAS's thread table (query 'M20x1.5'); fit: limit deviations of a size from KOMPAS's tolerance table (query '50H7', '30h6')"
    • changedInput schema / properties / kind / enum
      Previous value: -[
      -  "face",
      -  "edge",
      -  "material"
      -]New value: +[
      +  "face",
      +  "edge",
      +  "material",
      +  "thread",
      +  "fit"
      +]
    • changedInput schema / properties / query / description
      Previous value: -"Material search text (kind='material')"New value: +"material search text / thread / size+field"
  2. First observedv0.4.2

TDQS

A3.6/5.0
Behavior4/5

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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is free. The description adds two non-obvious behaviors that matter: returned face/edge results are 'topo:' handles valid only for the returned model revision, and expect_count violations produce a SELECTION_MISMATCH listing all candidates with attributes. It still doesn't say how multiple filters combine or what happens when nothing matches.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

It is short and front-loaded, leading with the geometric/reference-data split, and each sentence carries a distinct rule. The line breaks inside sentences make it slightly harder to scan, but there is essentially no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a zero-required, 10-parameter tool with no output schema, the definition covers the critical handshake (topo: validity, mismatch reporting) but leaves the return shape vaguely described and does not explain how the filter parameters interact or what an empty result looks like. Adequate, but gaps remain at this complexity level.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 60%, so the schema does partial work. The description adds real meaning for expect_count (failure mode and candidate listing) and reinforces the 'never by index' rule, but says nothing about normal, extreme, radius_mm, created_by, curve_type, or surface_type beyond the bare field names, leaving half the parameter space unexplained.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb (find/select) and disambiguates the five result kinds it can return: face/edge geometry, handbook materials, thread, and fit reference data. It is clear about what the tool retrieves, but it never distinguishes itself from sibling tools such as material_pick, which plausibly overlaps on the material/thread/fit path.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives useful operating rules: find geometry 'never by index', quote reference values 'from here, never from memory', and recommend expect_count. What it does not give is an explicit when-to-use-this-vs-another-tool statement, so selection relative to siblings like material_pick is left to inference.

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