What it does
This MCP server adds watermarks to AI-generated content in line with Article 50(2) of the EU AI Act and attaches a C2PA manifest envelope. Its listed features are AI content watermarking, detection and verification, batch processing and API integration. The exposed tools are generate_watermark, verify_watermark and attach_c2pa_manifest.
It suits workflows that produce or handle AI-generated text or media and need to mark those outputs or check whether a watermark is present. The publisher's documentation links it to Article 50 transparency obligations, and it is intended for use from an AI assistant that calls these tools.
The server is released under the MIT licence by MEOK AI Labs and is distributed as a Python package. It runs over stdio and is started from an MCP client configuration, such as Claude Desktop's config file. Once configured, the assistant calls the tools on request.
Limitations and requirements
Requires an MCP client that can be configured to launch the server over stdio. The publisher describes the Article 50 support as applicable only where relevant, so compliance depends on the user's own use case.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CSOAI-ORG/agent-content-watermark-mcp
Agent Content Watermark MCP — dedicated EU AI Act Article 50(2) GenAI watermarking. Visible +
v1.1.9As shown on registry.modelcontextprotocol.io when we last read it