Kennmarq.com

Regeln und Code zur Kennzeichnung von KI-Bildern, -Videos und -Audiodateien mit C2PA und Wasserzeichen, je nach Markt und Stack. Remote-Server ohne Schlüssel.

Gehosteter MCP-Server

npx add-mcp 'https://api.kennmarq.com/mcp'

Installiert in Claude Code, Codex, Cursor und mehr

Dokumentation

AI coding agents

Building with Cursor, Claude Code, Windsurf or Codex? Give your agent one prompt, or a standing rule, and it marks every image, video, audio file, 3D model and text your app generates for the EU AI Act and California. It reads our docs as plain text.

In short

  • The prompt: below, ready to paste.
  • Skill and MCP server: for Claude Code, Codex, Cursor, Copilot and Gemini CLI: they add the marks wherever your app makes AI media.
  • Docs for agents: kennmarq.com/llms.txt (the index), llms-full.txt (all of it), and any docs page as Markdown at its address plus .md.
  • The key: a free one from the dashboard, kept in KENNMARQ_API_KEY on the server.
  • Checking: drop a result on kennmarq.com/check.

1. Get a key

Sign up for free, confirm your email, and create a key in the dashboard: a test key while you build, a live one for production. Put it in your server's environment:

export KENNMARQ_API_KEY="sk_test_…"

2. Give your agent the prompt

Mark every AI image, video, audio file, 3D model and text this app generates with Kennmarq.com before it's stored or shown. The API docs are at https://kennmarq.com/llms.txt. Send each one to POST https://api.kennmarq.com/v1/mark (a text as JSON) with the key from the KENNMARQ_API_KEY environment variable, and pass provider, app_name, app_version, source and markets=eu,us-ca. For deep fakes, realistic depictions of real people, places or events, also send deepfake=true: it adds the visible "AI" label to a picture or video, and a spoken notice to a song. Keep only what comes back marked.

3. Or make it a standing rule

A rule keeps the marks in place as your app grows: every new feature that makes images, video, audio, 3D models or text gets them too. Add this to AGENTS.md, which many coding agents read, or to CLAUDE.md for Claude Code:

## AI content marking (Kennmarq.com)

- Every image, video, audio file, 3D model or text this app generates with AI is marked before it's stored or shown: send it to POST https://api.kennmarq.com/v1/mark and keep only what comes back.
- Call the API from server code only, with the key in the KENNMARQ_API_KEY environment variable. The key never reaches a browser or a mobile app.
- Send provider (our legal name), app_name and app_version (this app), source (created, composite or edited) and markets with the ISO codes of the countries the app serves (eu, us, jp…; see https://kennmarq.com/docs/laws).
- Deep fakes (realistic depictions of real people, places or events) also get deepfake=true, which draws the visible "AI" label into the picture or every frame of the video, and puts a spoken notice before a song's sound. label=true adds the label to any picture or video.
- A 3D model (GLB, glTF, PLY or OBJ) comes back with metadata naming it AI-generated: keep the .c2pa sidecar its X-Manifest-Url header names next to it.
- A text goes as JSON, with the text in "text", and comes back with a visible notice and Content Credentials inside it. In the EU, text over 200 tokens also needs the model's own watermark, declared with model_watermark.
- If marking fails, the request fails: never show or store an unmarked file.
- Files over 4 MB, and every video, go up first through POST https://api.kennmarq.com/v1/uploads. Videos are marked as jobs: 202, then GET the mark's url until it's completed.
- Docs: https://kennmarq.com/llms.txt; all of them in one file: https://kennmarq.com/llms-full.txt.

For Cursor, as a project rule in .cursor/rules/kennmarq.mdc:

---
description: Mark AI-generated images, video, audio, 3D models and text with Kennmarq.com
alwaysApply: true
---

- Every image, video, audio file, 3D model or text this app generates with AI is marked before it's stored or shown: send it to POST https://api.kennmarq.com/v1/mark and keep only what comes back.
- Call the API from server code only, with the key in the KENNMARQ_API_KEY environment variable. The key never reaches a browser or a mobile app.
- Send provider (our legal name), app_name and app_version (this app), source (created, composite or edited) and markets with the ISO codes of the countries the app serves (eu, us, jp…; see https://kennmarq.com/docs/laws).
- Deep fakes (realistic depictions of real people, places or events) also get deepfake=true, which draws the visible "AI" label into the picture or every frame of the video, and puts a spoken notice before a song's sound. label=true adds the label to any picture or video.
- A 3D model (GLB, glTF, PLY or OBJ) comes back with metadata naming it AI-generated: keep the .c2pa sidecar its X-Manifest-Url header names next to it.
- A text goes as JSON, with the text in "text", and comes back with a visible notice and Content Credentials inside it. In the EU, text over 200 tokens also needs the model's own watermark, declared with model_watermark.
- If marking fails, the request fails: never show or store an unmarked file.
- Files over 4 MB, and every video, go up first through POST https://api.kennmarq.com/v1/uploads. Videos are marked as jobs: 202, then GET the mark's url until it's completed.
- Docs: https://kennmarq.com/llms.txt; all of them in one file: https://kennmarq.com/llms-full.txt.

4. Or install the skill and the MCP server

Our agent kit holds a skill, which your agent loads whenever an app it builds generates or edits images, video, audio, 3D models or text, with the code for each stack: the Vercel AI SDK, JavaScript, Python, ComfyUI and plain HTTP. In Claude Code it comes as a plugin, with our MCP server:

/plugin marketplace add Kennmarq-com/kennmarq-agent-kit
/plugin install kennmarq@kennmarq

In any agent that reads skills (Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf and others):

npx skills add Kennmarq-com/kennmarq-agent-kit

The MCP server, at https://api.kennmarq.com/mcp, needs no key. Its tools answer what each market's labelling laws ask of a file and what a call must send (get_requirements), give the code for a stack (get_integration), and read any docs page (read_doc). Add it to Claude Code on its own, or to Cursor with one click:

claude mcp add --transport http kennmarq https://api.kennmarq.com/mcp

Claude Code reads AGENTS.md only when the project has no CLAUDE.md: put the rule in CLAUDE.md there, or add the line @AGENTS.md to it.

What your agent should send

FieldWhat it is
providerYour legal name, as the provider of the AI system.
app_name, app_versionYour app: the AI system California's latent disclosure names.
sourcecreated (made by AI), edited (an existing file changed with AI) or composite.
marketseu,us-ca, or the markets you serve: a call missing what their laws need is refused, and each file gets a receipt.
model (optional)The model's name, e.g. FLUX.1 schnell.

5. Check what it built

  • The key appears only in server code: search the project for KENNMARQ_API_KEY, and look for it in nothing a browser loads.
  • Every place that makes an image, a video or a song calls the API before storing or showing the file.
  • When marking fails, the request fails too: no unmarked file is shown or saved.
  • Files over 4 MB and videos go through an upload, and videos wait for their job (GET /v1/marks/{id}).
  • A marked image dropped on the checker shows your provider, your AI system and when it was made.

For the code an agent might write, see the Next.js and Vercel AI SDK guides.