PartReel API

All part data is served as static JSON. No auth, no fees, no rate limit — designed for AI agents to fetch directly.

Endpoints

• GET /api/v1/parts.json — full part list (id, name, keywords, detail API URL)
• GET /api/v1/parts/<id>.json — part detail (parameters, absolute download URLs, datasheet, verification status)
• GET /llms.txt — machine guide for AI agents

MCP server (AI agents / IDEs)

Remote MCP — nothing to install, register the URL: https://mcp.partreel.com/mcp

• Tools: search_parts(query) · get_part(id) · list_parts() · submit_part(...) — share a part instantly · report_feedback(part_id, result, notes) · request_part(...) · how_to_contribute()
• Claude Code: claude mcp add --transport http partreel https://mcp.partreel.com/mcp

Examples

/api/v1/parts.json · /api/v1/parts/usb_c_16p.json

Quality

Every part must pass automated gates (structure validation, KLC drawing rules, text-overlap, render completeness, STEP kernel checks). Dimensions are datasheet-derived and matched against the official KiCad library where one exists.

Add parts & feedback

Add parts and report field feedback via GitHub — CI quality gates auto-review contributions. Spec: CONTRIBUTING-AGENTS.md.

← Home