Integration

Very Good FFmpeg integration

Very Good FFmpeg runs your own FFmpeg command lines on hosted CPU (16 cores) or NVIDIA L4 GPU workers. Submit a job with input file URLs, output file names and one or more chained FFmpeg commands, wait for it or follow it later, list and cancel jobs, and get a one-hour presigned upload URL pair for a file that has no public URL (the upload itself is an HTTP PUT that no tool of this integration performs). Finished outputs come back as signed download URLs valid 7 days (files kept at least 30 days). Jobs are billed per GB processed (input read plus output written, failed and cancelled jobs included). The vendor's blocking wait mode is not exposed: the job tool waits by polling instead. Authentication is an API key from the Very Good FFmpeg dashboard, sent as a Bearer token.

Connect Very Good FFmpeg How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Very Good FFmpeg.Built from the 5 Very Good FFmpeg tools below, on the connection you add once.

What you can do with Very Good FFmpeg

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Job, 4 operations

cancel_jobCancel jobPOSTAgents: ask firstStops a queued or running job for good and returns it with status cancelled. Bytes already processed are still billed.
get_jobGet jobGETAgents: ask firstReturns one job: status (queued, running, succeeded, failed or cancelled), error_message, timings, byte totals and output_files, each output name mapped to a signed download URL valid 7 days.
list_jobsList jobsGETAgents: ask firstLists the account's FFmpeg jobs, each with its status, commands, input URLs, output download URLs and byte totals. Page with limit (1 to 200, default 50) and offset; pagingParams.hasMore says whether more remain.
run_ffmpeg_jobRun ffmpeg jobPOSTAgents: ask firstRuns FFmpeg commands on a hosted worker, billed per GB read plus written (failed jobs too). Inputs are URLs saved as inputs/<name>; outputs/<name> come back as signed URLs. Waits for the job and returns the finished result; gives up after 10 minutes.

Temp file, 1 operation

create_temp_file_uploadCreate temp file uploadPOSTAgents: ask firstReturns a fresh upload_url and download_url pair, both expiring after 1 hour, for a file with no public URL. The bytes go by HTTP PUT to upload_url, which no tool here does; download_url then goes in run_ffmpeg_job input_files.

Use Very Good FFmpeg from Claude Code, Cursor or Codex

Connect Very Good FFmpeg in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Very Good FFmpeg operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Very Good FFmpeg

Create your workspace
Sign up free. Nothing to install.
Connect Very Good FFmpeg once
Add your Very Good FFmpeg connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.