AI agents are getting good at reading code, writing documents, and navigating APIs. But hand them a 3D file — an STL from a manufacturing pipeline, a point cloud from a drone survey, a glTF from a game asset library — and they’re stuck. They can’t parse binary geometry formats, and bundling C++ geometry libraries into an agent is rarely practical.

Caliper fixes this. It’s a geometry file metadata API, built as an MCP server, that gives agents structured intelligence about any 3D file.

What Caliper does

Send a file (by URL or base64), get structured JSON metadata back:

• Bounding boxes — spatial extents in all three axes
• Mesh statistics — vertex count, face count, edge count
• Surface properties — surface area, volume (for watertight meshes)
• Topology checks — manifold status, watertight verification
• Point cloud stats — point count, density, classification distribution
• Scene structure — node hierarchy, materials, animations (glTF)

Six formats are supported today, with more coming:

Format	What Caliper extracts
STL (ASCII + binary)	Triangles, vertices, bounding box, surface area, volume, manifold/watertight checks
OBJ	Vertices, faces, materials, groups, bounding box
PLY (ASCII + binary)	Vertices, faces, vertex properties, bounding box
PCD	Points, fields, density, bounding box
LAS/LAZ	Points, classifications, bounds, coordinate system
glTF/GLB	Nodes, meshes, materials, animations, total vertices/primitives

STEP and IGES support is on the roadmap for CAD-native formats.

Why MCP?

The Model Context Protocol lets any MCP-compatible client — Claude, Cursor, custom agents — discover and call Caliper’s tools without custom integration code. Add the endpoint, and your agent has geometry awareness.

MCP endpoint: https://caliper.fit/mcp/

There’s also a REST API for applications that aren’t using MCP. Same tools, same metadata, plain HTTP.

How payment works

Caliper uses the x402 protocol for pay-per-call pricing in USDC on Base. No accounts, no API keys, no billing dashboards. Agents pay autonomously per request.

Format detection (caliper_detect_format) is free — useful for routing files before committing to a paid analysis.

Who is this for?

AI agent developers who need their agents to reason about 3D files without bundling geometry libraries. Instead of installing Open3D or trimesh and dealing with C++ dependencies, your agent calls Caliper and gets clean JSON.

CAD/manufacturing platform teams who need lightweight file triage — quick poly counts, bounding boxes, and manifold checks before committing to expensive processing.

Asset pipeline operators who need automated quality checks across libraries of 3D files.

GIS engineers working with LAS/LAZ point clouds who want quick metadata without downloading and processing multi-gigabyte files.

Try it now

The fastest way to start:

1. Free format detection — no payment needed:

   curl -X POST https://caliper.fit/tools/caliper_detect_format \
     -H "Content-Type: application/json" \
     -d '{"file_url": "https://raw.githubusercontent.com/mikedh/trimesh/main/models/featuretype.STL"}'

2. MCP setup — add https://caliper.fit/mcp/ to your MCP client and start calling tools.

3. Full guides — Getting Started with MCP | Getting Started with REST

What’s next

• STEP and IGES support — CAD-native formats for engineering workflows
• Integration guides — LangChain, CrewAI, AutoGen tutorials
• More metadata — material properties, coordinate system detection, file repair suggestions

We built Caliper because we believe geometry intelligence should be as easy for agents to access as text or images. Try it at caliper.fit and let us know what you build.

---

Caliper is live at caliper.fit. API docs | MCP manifest