QR Code Studio — Agent Integration (agents.md)

This page is the canonical agent onboarding guide for QR Code Studio. The same file lives at the repository root as agents.md.

Start here

Priority What Where
1 This guide /dev/docs.php?path=agents
2 REST API API reference · OpenAPI JSON
3 SMCP plugin SMCP docs · GitHub smcp_plugin · cli.py --describe
4 SDKs PHP (GitHub) · Python (GitHub)

What agents should use

Prefer the SMCP plugin when your host supports Sanctum MCP plugins — it wraps normalize + generate with --describe discovery.

Otherwise call the REST API directly or use the PHP/Python SDK in this repo.

Endpoints (summary)

GET  /api/v1/health.php
POST /api/v1/normalize.php   { "content": "...", "type": "auto" }
POST /api/v1/generate.php    { "content": "...", "format": "png", "ecl": "M" }
GET  /api/v1/openapi.json

Normalize example

curl -sS -X POST 'https://qr.decisionsciencecorp.com/api/v1/normalize.php' \
  -H 'Content-Type: application/json' \
  -d '{"content":"hello@gnail.com","type":"email"}'

Response includes encoded, type, and optional suggestion / suggestion_reason for typos.

Generate example

curl -sS -X POST 'https://qr.decisionsciencecorp.com/api/v1/generate.php' \
  -H 'Content-Type: application/json' \
  -d '{"content":"https://decisionsciencecorp.com","format":"png","ecl":"M"}'

Decode data_base64 to bytes for the image file.

Authentication

Public by default with IP rate limiting (~120 requests/minute). When the operator sets QR_STUDIO_API_KEY on the server, send:

X-API-Key: <key>

or

Authorization: Bearer <key>

Logo overlay (API)

Pass logo_base64 (raw base64 or data:image/png;base64,...) and logo_size_pct (10–30). The API bumps error correction to at least Q when a logo is present.

Content types

type Encoded payload
url https://... (scheme inferred if missing)
email mailto:user@domain
tel tel:+15550100123
text literal string

SMCP commands

Command Purpose
health Service health
normalize Validate / autocorrect
generate PNG or SVG as base64
tool-help Intent → recommended command

Run discovery:

python3 smcp_plugin/qr_studio/cli.py --describe

Source: smcp_plugin on GitHub

Licensing for agents

  • Code you execute from this repo (SDK, SMCP): AGPL-3.0 — network use requires source offer.
  • This documentation: CC BY-SA 4.0 — attribute Decision Science Corp.

Human UI

The public studio at / works offline in the browser. Agents should use /api/v1/ for deterministic automation.

Support