ffpipe — video conversion, probe & frames

Convert, probe, and split video into frames — real FFmpeg, paid per job with x402 on Base.

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

    What data it sees

    Do you need an account

    No: the server works without sign-in

    Convert, probe, and split video into frames — real FFmpeg, paid per job with x402 on Base.

    How to connect

    How to connect ffpipe — video conversion, probe & frames to Claude

    The server address to paste into Claude settings:

    https://ffpipe.dev/mcp

    1. Open Claude (claude.ai or the desktop app).
    2. Go to Settings → Connectors.
    3. Click Add custom connector.
    4. Paste the server address copied below into Remote MCP server URL and click Add.
    5. In a chat, click + → Connectors and switch the new connector on.

    Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

    Authorization

    After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

    How to connect ffpipe — video conversion, probe & frames to ChatGPT

    The server address to paste into ChatGPT settings:

    https://ffpipe.dev/mcp

    1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
    2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
    3. Open Settings → Connectors and click Create.
    4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
    5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
    6. In a new chat open + → Apps/Connectors and enable the connector.

    OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

    Authorization

    On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

    How to connect ffpipe — video conversion, probe & frames to Cursor

    The server address to paste into Cursor settings:

    https://ffpipe.dev/mcp

    Fastest: click Open in Cursor below and confirm the prompt.

    Manually:

    1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
    2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
    3. Save the file. The server appears in the MCP list; authorize it there if asked.

    Authorization

    If the service needs sign-in, an authorize button appears next to the server in the MCP list.

    Server tool list (6)

    Raw names from tools/list. Only developers need these.

    convert_videoQueue an asynchronous video conversion: fetches sourceUrl (or, with upload: true, waits for you to PUT the bytes to the returned uploadUrl), converts to MP4 scaled to fit maxWidth x maxHeight (defaults 1920x1080, cap 4096). Returns { jobId, status: "queued" | "awaiting_upload", statusUrl, uploadUrl?, uploadExpiresAt? } - poll with get_job_status (jobId + the t token from statusUrl) until finished/error. Download results within 24 hours. Paid: $0.02 USDC via x402 (base mainnet), or a convert-kind retryVoucher.
    probe_mediaQueue an asynchronous media inspection: fetches sourceUrl (or, with upload: true, waits for you to PUT the bytes to the returned uploadUrl) and returns its technical metadata as a downloadable file. In the default json format the result is the full raw analysis plus one added top-level "ffpipe_summary" object: container, duration_s, size_bytes, has_video, has_audio, video_codec, audio_codec, coded_width/height, display_width/height, rotation, fps, is_hdr, audio_channels, audio_sample_rate. USE display_width/display_height as the size the picture actually is - phones store portrait video as a landscape frame plus a rotation flag. A file that cannot be decoded is still a SUCCESS whose result carries the diagnosis. Returns { jobId, status, statusUrl, ... } - poll with get_job_status until finished/error, then download resultUrl within 24 hours. Paid: $0.01 USDC via x402 (base mainnet), or a probe-kind retryVoucher.
    extract_framesQueue an asynchronous frame extraction - the tool to reach for when you need to SEE a video: fetches sourceUrl (or, with upload: true, waits for you to PUT the bytes to the returned uploadUrl) and extracts still frames as jpeg, webp or png. mode is REQUIRED and picks the sampling (all, every_nth, interval, keyframes, evenly_spaced, timestamps); there is no default because every candidate implies a rate you did not ask for. A finished job's resultUrl is a JSON MANIFEST: source display dimensions, fps, rotation, an echo of your parameters, frame_count, and one entry per frame with index, pts_time (seconds in the source) and its own download url on this gateway - fetch the manifest, then the frames you want. Need just ONE frame at a moment - a thumbnail, poster frame or screenshot at a given second? Use mode "timestamps" with a single value, e.g. timestamps=[5], format=jpeg; the result is still a one-entry manifest pointing at that image (if you would rather have the image bytes returned directly instead of via a manifest, tell us with leave_feedback - demand decides whether we add it). Up to 1000 frames per job; "all"/"interval" on a long source can exceed that and fail fast, so prefer interval, evenly_spaced or timestamps. Unreadable input, audio-only files and still images FAIL here (probe_media diagnoses those). Download within 24 hours. Paid: $0.01 USDC via x402 (base mainnet), or a frames-kind retryVoucher.
    get_job_statusFree status lookup for a job accepted by convert_video / probe_media / extract_frames. Requires the job's ownership token - the `t` query parameter of the statusUrl returned at acceptance (store the whole statusUrl; it cannot be regenerated). Status goes [awaiting_upload ->] queued -> finished | error; awaiting_upload jobs report uploadExpiresAt but NEVER the uploadUrl (that credential is handed out exactly once, in the acceptance result); finished jobs carry resultUrl/thumbnailUrl/durationMs (served by this gateway, free to re-download within 24 hours of completion; GET them following redirects); a failed charged job carries retryVoucher. A bad token and an unknown jobId both answer not_found, deliberately.
    get_balanceFree. Reports the credit balance of the account this client is connected to, plus what one credit is worth, what each operation costs in credits, the monthly free allowance and whether a card is on file to unlock it. Takes no input: it reads the account behind the credential your client already connected with. If you have a connected account, paid tools are funded from this balance and never present a payment challenge - so call this to find out how much work you can still queue, and top up at the account page named in the answer. Anonymous callers (x402 buyers, who pay per call instead) have no balance and get an unauthorized error; either way nothing here is charged.
    leave_feedbackTell us what you needed. If ffpipe does not offer the operation you came for - another output format, extracting a frame, a different transformation entirely - say so here: these requests are aggregated and they drive directly what gets built next, so a one-line "I wanted X" is genuinely useful to us. Also the place for bug reports and any other feedback. Free, unauthenticated, and one-way: nothing is returned but an acknowledgement, and no job is created. Include jobId if a specific job prompted this (context only - it is not checked and is not required), and contact if you want a reply. Message limit 2000 characters.