Skip to main content
Glama
inventor-mcp-server

inventor-mcp-server

check_fastener_alignment

Read-only

Checks whether holes, slots, and cut-outs on mating flat faces in an Inventor assembly line up, reporting misaligned, blocked, or engaged openings before fabrication.

Instructions

Do the holes and slots of parts that lie on each other line up? Run it on an assembly before it goes to the shop — a hole that is 1.5 mm off costs a part.

Wherever a flat face of one part lies on a flat face of another (facing each other, at most max_gap apart), every hole, slot and cut-out in those faces is compared with what is behind it: aligned an opening of the other part, off by at most tolerance (counted, by size) misaligned an opening that is off: by how much, and the largest bolt that still passes engaged a pin, stud or boss of the other part goes into it (counted, by size); stud_does_not_fit lists the ones that are too big or too far off blocked solid material of the other part — a hole nothing can go through open the other part does not reach there Positions are in assembly mm. Fasteners (Content Center parts) and hidden components are left out. The geometry is read inside Inventor and kept per file on disk, so a 400-part assembly takes seconds, and the next call less; a part with unsaved changes is read anew. Changes nothing. Holes are compared as modelled in the part files: an assembly-level cut is not seen. Only flat faces are compared, and what sits INSIDE a hole is not looked at: an "open" hole may hold a pin or a rivet nut that is its own part. Each row has at (assembly mm) and axis (out of the part, towards the other one) — capture_view(look_from=…, target=at) shows the spot.

Args: parts: Name patterns, ","-separated, on the component path or the file name ("DOOR,ET1-0?.ipt"); empty = every part. max_gap: Faces up to this far apart (mm) count as lying on each other — a gasket or powder coat gap; 0.5 by default. tolerance: Openings off by more than this (mm) are misaligned. refresh: Read every file again instead of using what is kept on disk. include_fasteners: Also compare Content Center parts. limit: Rows listed per kind; the counts are always complete.

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": "check_fastener_alignment"
  }
}

Input Schema

NameRequiredDescriptionDefault
limitNo
partsNo
max_gapNo
refreshNo
instanceNoWhich connected Inventor to use (list_instances). Leave empty when only one is connected.
toleranceNo
include_fastenersNo

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.2.0

TDQS

A4.3/5.0
Behavior5/5

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description still adds substantial behavior: geometry is read inside Inventor and cached per file on disk (a 400-part assembly takes seconds, the next call less), unsaved parts are re-read, refresh forces a full re-read, and nothing is modified. It also discloses real limitations an agent must know: assembly-level cuts are not seen, only flat faces are compared, and what sits INSIDE a hole is not inspected, so an "open" hole may still hold a pin. Each row's `at` and `axis` fields are explained and linked to capture_view.

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?

The description is front-loaded with the question and the shop-floor motivation, then moves to a classification legend and an Args block — a structure an agent can scan. The five-kind legend is long and, since an output schema exists, partly restates return values rather than adding invocation knowledge. Overall it is dense but nearly every clause carries operational meaning, with only mild redundancy in the legend.

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

Completeness5/5

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

For a 7-parameter analysis tool with sparse schema descriptions, the definition supplies the trigger, the comparison rules, the caching/refresh model, the exclusion list, the known blind spots, and the row shape with a pointer to capture_view for visual verification. Nothing an agent needs in order to call it correctly and interpret blind spots is missing, and the output schema is left to describe return structure.

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

Parameters4/5

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

Schema description coverage is only 14%, so the description must carry the semantics and it largely does: `parts` gets pattern syntax and a concrete example ("*DOOR*,ET1-0?.ipt") plus the empty-string meaning, `max_gap` is explained as a gasket/powder-coat gap with its 0.5 default, `tolerance` as the misalignment threshold, `refresh` as bypassing the disk cache, `include_fasteners` as opting Content Center parts in, and `limit` as rows-per-kind while counts stay complete. The only parameter not touched is `instance`, which the schema itself documents. Deducting one point because defaults for `tolerance` and `limit` are left to the schema rather than restated where the prose discusses them.

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 opens with a concrete, specific question ("Do the holes and slots of parts that lie on each other line up?") and then defines exactly what is compared: openings in coincident flat faces, classified as aligned/misaligned/engaged/blocked/open. That is far beyond a restatement of the name and leaves no ambiguity about the operation. It never names a sibling alternative (e.g. check_interference) for the closely related job of checking part-to-part problems, so it falls short of full differentiation.

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

Usage Guidelines4/5

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

"Run it on an assembly before it goes to the shop — a hole that is 1.5 mm off costs a part" gives a clear situational trigger that tells the agent when this tool is worth calling. Scope limits also act as usage guidance: fasteners and hidden components are skipped by default, only flat-face contacts are judged, and assembly-level cuts are invisible. No explicit when-not or named alternative tool is provided, so the routing burden beyond the stated trigger is left to the agent.

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