video-frame

Video frame extraction tool: captures a single frame from the video by mode (first, last, or a given time) and returns the uploaded image URL.

Overview

Field	Value
Model ID	`video-frame`
CLI	`dlazy video-frame`
MCP	`video-frame` (Claude Code surfaces this as `mcp__dlazy__video-frame`)
Type	`image`
Execution	Returns synchronously
Batch	Supports `--batch <n>` parallel fan-out

Parameters

Arg	Type	Required	Notes
`file`	url	Yes	File URL; video (accepts URL, local path, or data: URL)
`mode`	"first" \| "last" \| "current"	Yes	Mode
`time`	string	No	Time (s)

--input @file.json or --input '{...}' can supply all args at once; flags take precedence over --input keys.

CLI Examples

dlazy video-frame --help
dlazy video-frame --file "https://example.com/source.mp4" --mode "first"
dlazy video-frame --file "./local-video.mp4" --mode "first"
dlazy video-frame --file "https://example.com/source.mp4" --mode "first" --dry-run
dlazy video-frame --file "https://example.com/source.mp4" --mode "first" --batch 4

Compose with a pipeline

dlazy seedance-2.0-fast --prompt "reference clip" \
  | dlazy video-frame --file -

MCP is consumed by AI clients, not handwritten. Once dLazy is registered as an MCP server the tool appears in the client's tool list — Claude Code surfaces it as mcp__dlazy__video-frame; generic clients (e.g. OpenClaw) call it as video-frame.

See MCP setup for how to add the server.

Output

{
  "outputs": [
    { "type": "json", "id": "o_...", "value": { /* tool-specific payload */ } }
  ]
}

Structured payloads are returned as json outputs; the inner shape is tool-specific.