PicDefense.io MCP Server
Run PicDefense.io image risk and content analysis on any public image URL through 11 MCP tools.
Check account credits — see your remaining PicDefense.io balance before making billable calls.
Assess image risk — core reverse-image-search analysis with a "picrisk" score showing how widely/riskily an image appears online.
Extract metadata — pull EXIF (camera, timestamps, GPS) plus IPTC/XMP copyright, creator, credit and licensor fields.
Detect faces — identify whether a human face is present.
Detect landmarks — recognize landmarks and their location in the image.
Detect logos — identify brand logos.
Run content safety checks — SafeSearch assessment (adult, violence, racy, medical, spoof likelihoods).
Find backlinks — locate other web pages where the image appears.
Detect labels — describe image contents (objects, scenes, concepts).
Extract text via OCR — return full text, individual words, and a truncation flag.
Detect watermarks — spot visible stock/photographer watermarks with confidence and identified source.
All image tools accept a single public http/https image URL and most consume account credits; the server is multi-tenant, forwarding each caller's USERID:APIKEY token as X-API-Token, and supports Streamable HTTP, legacy SSE, and local stdio transports.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@PicDefense.io MCP Server check risk of https://example.com/photo.jpg"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
PicDefense.io MCP Server
A Model Context Protocol server for the PicDefense.io API — let AI agents run reverse-image risk analysis, EXIF/IPTC/XMP metadata extraction, image backlink discovery, and image content detection (face / landmark / logo / label / SafeSearch) on any image URL.
Features
🔐 Per-user authentication — each connection carries its own PicDefense API token; the server holds no keys
🚀 Dual transport — modern Streamable HTTP (
/mcp) and legacy SSE (/sse)🧰 11 tools covering the full PicDefense API v2
🐳 Docker-ready — production container behind nginx
📖 Built-in docs — Swagger UI at
/docs
Related MCP server: MCP Image Extractor
Tools
Tool | Description |
| Remaining account credit balance |
| Reverse-image risk analysis + picrisk score (core tool) |
| Extract EXIF metadata (camera, timestamps, GPS) plus IPTC/XMP copyright, creator, credit and licensor fields |
| Detect a human face in an image |
| Detect a recognizable landmark |
| Detect a brand logo |
| Content-safety (adult/violence/racy/…) assessment |
| Find pages where an image appears |
| Detect descriptive labels for image contents |
| Extract text from an image via OCR |
| Detect a visible stock/photographer watermark (source + confidence) |
All image tools take a single url (a public http/https image URL). Most tools
consume account credits per call — use picdefense_get_credits to check your balance.
Authentication
Every request authenticates with your PicDefense API token, which is your user id and API key joined by a colon:
USERID:APIKEYFind both in your PicDefense.io account settings: https://app.picdefense.io/?returnUrl=https://app.picdefense.io/dashboard/settings
The token is sent as the X-API-TOKEN header to the API (https://app.picdefense.io/api/v2).
Quick start
Hosted server (recommended)
The hosted server runs at https://mcp.picdefense.io. Add it to Claude Code:
# Streamable HTTP (recommended)
claude mcp add -t http picdefense "https://mcp.picdefense.io/mcp" \
--header "X-API-Token: USERID:APIKEY"
# or SSE
claude mcp add -t sse picdefense "https://mcp.picdefense.io/sse?token=USERID:APIKEY"Quick HTTP smoke test:
curl -X POST https://mcp.picdefense.io/mcp \
-H "Content-Type: application/json" \
-H "X-API-Token: USERID:APIKEY" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'Claude Desktop (hosted)
Claude Desktop launches MCP servers as local commands, so reach the hosted server
through the mcp-remote bridge (requires
Node.js installed). See claude_desktop_config.example.json:
{
"mcpServers": {
"picdefense": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://mcp.picdefense.io/sse?token=YOUR_USERID:YOUR_APIKEY"
]
}
}
}Edit your claude_desktop_config.json (Settings → Developer → Edit Config), add the
mcpServers block above with your USERID:APIKEY, then fully quit and reopen Claude Desktop.
Testing against a plain-HTTP server (e.g.
http://<host>:6910) instead of HTTPS?mcp-remoteblocks non-HTTPS origins unless the host islocalhost— append"--allow-http"to theargsarray, or reach it over an SSH tunnel tolocalhost.
Local (stdio) via npx — no clone needed
Run the published package directly. Requires Node.js installed.
{
"mcpServers": {
"picdefense": {
"command": "npx",
"args": [
"-y",
"@picdefenseio/mcp-server",
"--api-token",
"USERID:APIKEY"
]
}
}
}You can also pass the token via the PICDEFENSE_API_TOKEN env var instead of --api-token.
Local (stdio) from source
Clone and build, then point Claude Desktop at the built entry point:
git clone https://github.com/rchanllc/picdefenseio-mcp-server.git
cd picdefenseio-mcp-server
npm install
npm run buildThen use "command": "node" with "args": ["/absolute/path/to/dist/index.js", "--api-token", "USERID:APIKEY"].
Configuration
Variable | Default | Description |
| — |
|
|
| API base URL |
|
| Hosted server listen port |
Running the hosted server
Development
npm run dev:sse # tsx watch, auto-reloadProduction (Docker)
# via docker compose
docker compose up --build -d
# or the helper script (handles build + health check)
./deploy.shThe container is named picdefenseio_mcp and listens on port 6910.
HTTP endpoints
Method | Path | Purpose |
GET |
| Health check (used by Docker + CI) |
GET |
| Server + transport info |
GET |
| List available tools |
GET |
| Swagger UI for the underlying API |
GET |
| Open an SSE MCP session |
POST |
| SSE session message channel |
POST |
| Streamable HTTP MCP (header |
Self-hosting
There is no CI/CD in this repo — host it yourself. On your server:
git clone https://github.com/rchanllc/picdefenseio-mcp-server.git
cd picdefenseio-mcp-server
docker compose up --build -d # or: ./deploy.shThe container is named picdefenseio_mcp and listens on 6910. To update,
git pull and re-run docker compose up --build -d.
Front it with nginx at https://mcp.picdefense.io → 127.0.0.1:6910
(proxy_buffering off and a long read timeout are recommended for the /sse path).
Architecture
┌──────────────┐ ┌────────────────────────┐ ┌─────────────────────────┐
│ MCP Client │───▶│ PicDefense MCP Server │───▶│ PicDefense.io API v2 │
│ (Claude etc.)│ │ (port 6910) │ │ app.picdefense.io/api/v2│
└──────────────┘ └────────────────────────┘ └─────────────────────────┘
token (USERID:APIKEY) forwarded as X-API-TOKEN ───────────▶Each connection builds its own API client + MCP server bound to the caller's token, so the service is multi-tenant and stateless with respect to credentials.
License
MIT — see LICENSE.
Available Tools
11 toolspicdefense_check_image_riskAInspect
Run PicDefense's core image-risk analysis on an image URL. Returns reverse-image-search findings and a "picrisk" score indicating how widely / riskily the image appears online.
This is the primary tool for assessing whether an image has been misused or scraped. Consumes account credits.
Inputs: url (required).
picdefense_detect_faceAInspect
Detect whether an image contains a human face. Consumes account credits.
Inputs: url (required).
picdefense_detect_labelsAInspect
Detect descriptive labels for the contents of an image (objects, scenes, concepts). Consumes account credits.
Inputs: url (required).
picdefense_detect_landmarkAInspect
Detect whether an image contains a recognizable landmark (and where). Consumes account credits.
Inputs: url (required).
picdefense_detect_logoAInspect
Detect whether an image contains a brand logo. Consumes account credits.
Inputs: url (required).
picdefense_detect_watermarkAInspect
Detect a visible stock/photographer watermark in an image. Returns whether one was detected, a confidence score, and the identified source (e.g. a stock agency, "photographer", or "copyright"). Consumes account credits.
Inputs: url (required).
picdefense_extract_exifAInspect
Extract EXIF metadata (camera, timestamps, GPS, etc.) embedded in an image at the given URL. Consumes account credits.
Inputs: url (required).
picdefense_extract_textAInspect
Extract text from an image via OCR. Returns the detected full text, individual words, and a "truncated" flag (text is capped for very text-dense images). Consumes account credits.
Inputs: url (required).
picdefense_find_backlinksAInspect
Find backlinks for an image — other web pages where the image appears. Useful for tracing where a photo has been republished. Consumes account credits.
Inputs: url (required).
picdefense_get_creditsAInspect
Get the remaining credit balance on your PicDefense.io account.
Most analysis tools consume credits per call — check this first if calls start failing with "insufficient credits".
Inputs: no named parameters.
picdefense_safesearchAInspect
Assess the content safety of an image (adult, violence, racy, medical, spoof likelihoods) via SafeSearch. Consumes account credits.
Inputs: url (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 11 tools
Each tool targets a distinct aspect of image analysis: risk, EXIF, faces, landmarks, logos, content safety, backlinks, labels, OCR, watermark, and credits. The descriptions clearly delineate overlapping concepts like reverse-image-search findings vs. dedicated detection tools, so an agent can unambiguously select the right one.
All tool names follow the 'picdefense_<verb_noun>' pattern (e.g., extract_exif, detect_face, find_backlinks) with a consistent prefix, but 'safesearch' deviates as a single compound noun rather than a verb_noun construction. This minor inconsistency does not hinder readability.
With 11 tools, the server achieves a well-scoped balance—enough to cover the full spectrum of image risk and detection features without overwhelming the agent. Each tool earns its place and there is no apparent bloat.
The tool surface comprehensively covers the image-risk domain: credit management, core risk scoring, EXIF, face/landmark/logo/watermark detection, content safety, backlinks, labels, and OCR. No obvious lifecycle or analytical gaps exist for typical use cases.
Maintenance
Related MCP Connectors
A Model Context Protocol server for Wix AI tools
MCP server exposing the YourImageShare upload API (upload/list/delete) as tools for AI agents.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Create images & video from any MCP agent — 17 models, spend limits, one URL.
Related MCP Servers
- AlicenseAqualityDmaintenanceModel Context Protocol server that enables Claude Desktop (or any MCP client) to fetch web content and process images appropriately.1MIT
- AlicenseAqualityDmaintenanceA Model Context Protocol server that extracts images from URLs or base64 data and converts them into a format suitable for LLM analysis, allowing AI models to process and understand visual content.3386 npm21MIT
- AlicenseAqualityBmaintenanceA Model Context Protocol server that enables fetching and processing images from URLs, local file paths, and numpy arrays, returning them as base64-encoded strings with proper MIME types.122MIT
- AlicenseNot gradedqualityDmaintenanceProvides image analysis, inspection, cropping, OCR, and comparison capabilities via the Model Context Protocol, allowing AI agents to process and manipulate images using vision models.69 PyPI2MIT