
Security News
Deno Goes the Way of the Dinosaur
Cloudflare is acquiring Deno’s team, which will provide another year of bug fixes and security updates before ending support for the runtime.
vid2llm-mcp
Advanced tools
Lets AI agents read public videos (TikTok, YouTube, Instagram, X) as text through vid2llm.dev, paying per video with x402. MCP server and CLI.
Lets an AI agent read videos. Give it a TikTok, YouTube, Instagram or X link and it gets the author, description, stats and a timestamped transcript as text, through vid2llm.dev.
Each video is paid for individually with x402 in USDC on Base, from one cent. There is no account and no API key: the agent pays from a wallet you give it.
Claude Code:
claude mcp add vid2llm -e VID2LLM_PRIVATE_KEY=0x... -- npx -y vid2llm-mcp
Any other MCP client:
{
"mcpServers": {
"vid2llm": {
"command": "npx",
"args": ["-y", "vid2llm-mcp"],
"env": { "VID2LLM_PRIVATE_KEY": "0x..." }
}
}
}
The same package is a command-line tool, for agents that have a shell:
npx -y vid2llm-mcp price <video-url> # free: length and cost
npx -y vid2llm-mcp watch <video-url> # paid: markdown on stdout
npx -y vid2llm-mcp watch <video-url> --visuals # also what is shown on screen
watch needs VID2LLM_PRIVATE_KEY in the environment. Exit code 0 means the video was read.
VID2LLM_PRIVATE_KEY is the private key of the wallet that pays. Use a dedicated wallet holding
a few dollars of USDC on Base, not your main one. It needs no ETH; the payment facilitator
covers gas. The key stays on your machine and is only used to sign each payment.
watch_video(url, visuals?) returns the video as markdown. visuals: true adds on-screen
text and a description of what is shown, and costs more.video_price(url) is free and returns the video's length and what it would cost.The server also tells the agent when to use these, so it reaches for watch_video when you
share a video link.
| Variable | Default | Meaning |
|---|---|---|
VID2LLM_PRIVATE_KEY | Key of the paying wallet | |
VID2LLM_API_TOKEN | An access token, if you have been given one. Used instead of paying | |
VID2LLM_MAX_PRICE | $1 | The most it will pay for one video, whatever the server asks |
VID2LLM_URL | https://vid2llm.dev | The service to call |
Priced by the video's length, per started minute: $0.005 plus $0.005 a minute for the transcript, so a clip under a minute is one cent and a ten-minute video is about six. Nothing is charged when a video can't be read. Current rates: https://vid2llm.dev/pricing
FAQs
Lets AI agents read public videos (TikTok, YouTube, Instagram, X) as text through vid2llm.dev, paying per video with x402. MCP server and CLI.
The npm package vid2llm-mcp receives a total of 529 weekly downloads. As such, vid2llm-mcp popularity was classified as not popular.
We found that vid2llm-mcp demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.

Security News
Cloudflare is acquiring Deno’s team, which will provide another year of bug fixes and security updates before ending support for the runtime.

Security News
Anthropic’s free OSS Scanner uses Mythos and other Claude models to find open source vulnerabilities and send maintainers reports with proposed fixes.

Security News
On Front Lines’ Early Adopters: CISO podcast, Andrew Becherer shares lessons from Datadog, weighs the long-term costs of AI-built tools, and explains how founders can actually gain a CISO’s trust.