Platform
MCP & Skills for LLMs
Give an AI agent one tool that checks whether an image, a video or an audio recording is real, and a Skill that tells it what to do with the answer.
New accounts start with 100 credits.
- One detection tool for images, video and audio
- A verdict, a score and a summary, with per-frame or per-window evidence
- Hosted at mcp.scam.ai, or run it yourself with npx
Connect in one step
Add the server to your agent's MCP configuration. Hosted, there is nothing to install: point the client at https://mcp.scam.ai/mcp and send your API key in an x-api-key header. On your own machine, run npx -y @scam-ai/mcp-server with the key in SCAMAI_API_KEY; it needs Node 22 and is the one that can read a local file path.
Both reach the same detection, use the same key and cost the same as a direct API call.
Hosted
Nothing to install. Works in clients that cannot run a local process.
{ "mcpServers": { "scamai": { "url": "https://mcp.scam.ai/mcp", "headers": { "x-api-key": "YOUR_SCAMAI_API_KEY" } } }}
Run it yourself
Runs on your machine, so it can read a local file. Needs Node 22.
{ "mcpServers": { "scamai": { "command": "npx", "args": ["-y", "@scam-ai/mcp-server"], "env": { "SCAMAI_API_KEY": "YOUR_SCAMAI_API_KEY" } } }}
Paste a key from API keys in place of YOUR_SCAMAI_API_KEY.
One tool, and two free calls
detect_media checks one image, video or audio file, sent as a link, a local file path or the file's bytes, and works out which kind it is. get_credit_balance returns what the account has left, and get_usage lists recent credit movements; both are free.
A check is never retried for you: a retried check runs, and is billed, again.
detect_mediaThe media's price- Checks one image, video or audio file and works out which kind it is. Takes one of
url,file_pathorfile_base64. get_credit_balanceFree- What the key's account has left. Takes no arguments.
get_usageFree- Recent credit movements, newest first. After a timeout, read it before sending the same file again. Takes an optional
limit.
What comes back
Every answer carries a verdict to branch on. LIKELY_REAL means let it through, ALERT means put a human on it, and LIKELY_AI means block or escalate; audio answers LIKELY_REAL or LIKELY_AI only.
The score is the probability, from 0 to 1, that the media is AI-made or manipulated. It is not a confidence, and it is not proof. A one-sentence summary rides with it, and video adds a per-frame series and audio a per-window timeline, so a reviewer can see where in the clip the signal sits.
Every field is explained in Reading the result.
{ "status": "completed", "result": { "id": "det_4c8e2b91", "status": "completed", "created_at": "2026-09-15T10:02:41.517Z", "media": { "type": "video", "filename": "clip.mp4", "mime_type": "video/mp4", "bytes": 8123904 }, "model": "Eva V1.6", "verdict": "ALERT", "score": 0.57, "summary": "We detected some AI-related signals, but the result is not conclusive.", "credits_used": 1, "threshold_used": 0.5, "frames": [ { "index": 0, "timestamp_s": 0.0, "score": 0.5731 }, { "index": 1, "timestamp_s": 1.0, "score": null, "cause": "no_face" } ] }}
Skills: what the agent does with the answer
The tool answers one question per file. A Skill is the short instruction file that tells the agent what to do with each answer: branch on the verdict, send anything that is not clearly real to a person, and never present a score as proof. Written into a Skill rather than a prompt, the rule holds however far the conversation drifts.
The Skill on this page is a starting point: copy it into your agent's skills folder and adapt its rules to your own review policy.
---name: verify-media-before-actingdescription: Before relying on an image, video or audio file in a task, check it with the Scam AI MCP server and route anything not clearly real to a person.--- # Verify media before acting Call the scamai server's detect_media tool on every image, video or audio file the task depends on, before you summarize it, forward it or recommend an action based on it. 1. Call detect_media once per file, with url, file_path or file_base64. Keep the returned id with the file.2. Branch on verdict, never on your own reading of the file: - LIKELY_REAL: continue. - ALERT: do not rely on the file. List it for human review with its score and summary. - LIKELY_AI: do not act on the file. Escalate it with its score and summary.3. Report the score as the probability that the file is AI-made or manipulated. It is not proof, and it says nothing about a person.4. A failed call is not a verdict. Record error.code. Do not resend a file that returned unprocessable_media. After a timeout, call get_usage before sending the same file again: the check may have run and been charged.5. End with the result for each file and the list of files that need a person. The person decides; never approve, deny or accuse.
What it costs
The same as a direct API call, billed in credits to the account behind the key, and a failed check that could not be judged is never charged. Document forgery detection is sold on its own plan.
One credit per image, per scored video frame (at most 20 a clip) and per started minute of audio, at $0.02 a credit. Credits never expire, and new accounts start with 100. Full pricing
Where it fits
Anywhere an agent is about to rely on media that someone else supplied: the photos and recorded statement in an insurance claim, a selfie or ID video at onboarding or in a loan application, a candidate's interview recording, or a reported post in a trust and safety queue.
Common questions
What is an MCP server for deepfake detection?
MCP, the Model Context Protocol, is an open standard that lets an LLM or an agent call real tools. Scam AI's MCP server exposes one detection tool, detect_media: the agent passes an image, a video or an audio recording, as a link, a file path or the file's bytes, and gets back a verdict, a score from 0 to 1 and a one-sentence summary, with per-frame scores for video and a per-window timeline for audio. It answers exactly as the Scam AI API does and is billed the same way.
How do I use a deepfake-detection MCP server in an AI agent?
Add the server to your agent's MCP configuration: the hosted URL, https://mcp.scam.ai/mcp, with your API key in an x-api-key header, or npx -y @scam-ai/mcp-server with the key in SCAMAI_API_KEY. Then give the agent one rule per verdict: continue on LIKELY_REAL, send ALERT to a person, and block or escalate LIKELY_AI. A Skill is the place to write that rule, so it holds however the conversation drifts.
Which files can the Scam AI MCP server check?
Images (JPG, PNG or WEBP, up to 10 MB), video (MP4, MOV or WEBM, up to 100 MB) and audio (MP3, WAV, M4A, FLAC, OGG or AAC, up to 50 MB), or a link to a public video page or a media file. Document forgery detection, such as an edited PDF, is sold on its own plan and is not part of this tool.
Which LLMs and agent frameworks work with Scam AI's MCP server?
Any client that speaks MCP, including Claude, Cursor and VS Code. The hosted server suits a client that cannot run a local process; the npx server runs on your own machine and can read a local file. For a platform with its own tool format, the same detection is one REST call, described by a public OpenAPI file at https://api.scam.ai/openapi.json.
Connect
Connect your agent.
Create an account, paste one config and ask your agent about a file. New accounts start with 100 credits.