Video

Active

Your agent can't run media tooling or render video itself; this edits existing clips and generates new ones from a prompt.

4 tools

A complete video toolkit: edit existing footage (transcode, clip, concatenate, overlay, burn in subtitles, draw text) and generate short videos with audio from a text prompt in a regular and a high-quality tier. One call, MP4/WebM/GIF and more out.

Audio & Videovideotranscodeconcatoverlaysubtitlestextwatermarklogoconvertaudiomp4webmgifmp3

Tools (4)

Edit

▶transcode

Transcode, clip, concatenate, overlay, burn subtitles, or add text labels to video or audio.

Usage-based · 1 credit per second of output

Example prompts

“Convert this MP4 to WebM for use in a web player”
“Extract the audio track from this video as an MP3 at 192k bitrate”
“Clip seconds 10 to 40 from this video and export as MP4”
“Concatenate these three clips into one MP4”
“Burn these subtitles into the video and export as MP4”
“Overlay this logo PNG in the top-right of the video from seconds 2 to 8”
“Add a 'LIVE' label in 64px red text with a black background box in the top-left of this video”

Parameters

textobjectoptional

Burn a single text label into the video, optionally with a background box and time window.

concatarrayoptional

Additional clip URLs to concatenate after input_url. Output duration = sum of all clips. Up to 20 extra clips.

overlayobjectoptional

Overlay another video or image on top of the primary input.

end_timenumberoptional

Clip end time in seconds. Single-input mode only, ignored if concat is set.

input_urlstringrequired

Pre-signed GET URL of the input video or audio file. Obtain via POST /uploads/presign on faro-api.

video_crfintegeroptional

Video quality (CRF). 0=lossless, 18=visually lossless, 23=default H.264, 28=lower quality. Only applies to H.264 (mp4/mkv/avi) output.

expires_inintegeroptionaldefault: 3600

Download URL TTL in seconds (default 1h, max 24h).

resolutionstringoptional

Output resolution, e.g. '1280x720'. Use '-2' for the dimension you want auto-scaled, e.g. '1280:-2' keeps the aspect ratio.

start_timenumberoptional

Clip start time in seconds.

audio_bitratestringoptional

Audio bitrate, e.g. '128k', '192k', '320k'.

output_formatstringrequired

Target container/format. Use mp3/aac/wav/flac/ogg to extract audio only.

subtitles_urlstringoptional

Pre-signed GET URL of an SRT or VTT subtitle file to burn into the video.

output_filenamestringoptional

Filename for the output file. Defaults to `output.<ext>`.

API Usage

curl -X POST "https://skill.askfaro.com/skills/video/run" \
  -H "Authorization: Bearer faro_<your_key>" \
  -H "Content-Type: application/json" \
  -d '{
  "intent": {
    "prompt": "Convert this MP4 to WebM for use in a web player"
  }
}'

CLI Usage

askfaro describe video/transcode

Install pip install askfaro-cli, then askfaro auth login.

Generate

▶generate

Generate a short video with audio from a text prompt at a balanced, standard quality, up to 4K. If the response comes back with a continuation_token, call again with only that token until the video is ready.

Usage-based · 125 credits per second at 720p, 150 at 1080p, 375 at 4K

Example prompts

“Generate an 8-second cinematic clip of a fox running through a snowy forest at dawn”
“Create a 1080p time-lapse of clouds rolling over a mountain range, 6 seconds”

Parameters

modelstringoptionaldefault: "veo-3.1-fast-generate-preview"

Standard, balanced-quality video model.

promptstringoptional

Text prompt describing the video to generate. Required on the first call; ignored when continuation_token is set.

resolutionstringoptionaldefault: "720p"

Output resolution. 1080p and 4K force duration to 8 seconds.

aspect_ratiostringoptionaldefault: "16:9"

duration_secondsintegeroptionaldefault: 8

Clip length in seconds. Must be 8 when resolution is 1080p or 4k.

continuation_tokenstringoptional

Token from a prior pending response. When set, all other params are ignored and the server resumes polling; call again with only this token until the video is ready.

API Usage

curl -X POST "https://skill.askfaro.com/skills/video/run" \
  -H "Authorization: Bearer faro_<your_key>" \
  -H "Content-Type: application/json" \
  -d '{
  "intent": {
    "prompt": "Generate an 8-second cinematic clip of a fox running through a snowy forest at dawn"
  }
}'

CLI Usage

askfaro describe video/generate

Install pip install askfaro-cli, then askfaro auth login.

Generate (Fast)

▶generate_fast

Generate a short video with audio from a text prompt fast and at the lowest cost, up to 1080p. If the response comes back with a continuation_token, call again with only that token until the video is ready.

Usage-based · 62.5 credits per second at 720p, 100 at 1080p

Example prompts

“Generate a low-cost 6-second 720p clip of waves crashing on a beach”
“Make a quick 9:16 vertical clip of clouds rolling over a mountain range”

Parameters

modelstringoptionaldefault: "veo-3.1-lite-generate-preview"

Most affordable video model.

promptstringoptional

Text prompt describing the video to generate. Required on the first call; ignored when continuation_token is set.

resolutionstringoptionaldefault: "720p"

Output resolution. 1080p forces duration to 8 seconds.

aspect_ratiostringoptionaldefault: "16:9"

duration_secondsintegeroptionaldefault: 8

Clip length in seconds. Must be 8 when resolution is 1080p or 4k.

continuation_tokenstringoptional

Token from a prior pending response. When set, all other params are ignored and the server resumes polling; call again with only this token until the video is ready.

API Usage

curl -X POST "https://skill.askfaro.com/skills/video/run" \
  -H "Authorization: Bearer faro_<your_key>" \
  -H "Content-Type: application/json" \
  -d '{
  "intent": {
    "prompt": "Generate a low-cost 6-second 720p clip of waves crashing on a beach"
  }
}'

CLI Usage

askfaro describe video/generate_fast

Install pip install askfaro-cli, then askfaro auth login.

Generate (HQ)

▶generate_hq

Generate a short video with audio from a text prompt at the highest quality, up to 4K. If the response comes back with a continuation_token, call again with only that token until the video is ready.

Usage-based · 500 credits per second at 720p or 1080p, 750 at 4K

Example prompts

“Generate a high-quality 8-second 4K clip of a fox running through a snowy forest at dawn”
“Create a cinematic 1080p product teaser, 8 seconds, at the highest quality”

Parameters

modelstringoptionaldefault: "veo-3.1-generate-preview"

Highest-quality video model.

promptstringoptional

Text prompt describing the video to generate. Required on the first call; ignored when continuation_token is set.

resolutionstringoptionaldefault: "720p"

Output resolution. 1080p/4k force duration_seconds=8. Lite does not support 4k.

aspect_ratiostringoptionaldefault: "16:9"

duration_secondsintegeroptionaldefault: 8

Clip length in seconds. Must be 8 when resolution is 1080p or 4k.

continuation_tokenstringoptional

Token from a prior pending response. When set, all other params are ignored and the server resumes polling; call again with only this token until the video is ready.

API Usage

curl -X POST "https://skill.askfaro.com/skills/video/run" \
  -H "Authorization: Bearer faro_<your_key>" \
  -H "Content-Type: application/json" \
  -d '{
  "intent": {
    "prompt": "Generate a high-quality 8-second 4K clip of a fox running through a snowy forest at dawn"
  }
}'

CLI Usage

askfaro describe video/generate_hq

Install pip install askfaro-cli, then askfaro auth login.

▶README

Video Editor

Transcode, clip, concatenate, overlay, and burn subtitles into video or audio, all in a single call.

Workflow

Call POST /uploads/presign on faro-api to get a put_url + get_url pair for each input you need (primary, extra clips, overlay, subtitle file).
PUT your files to the put_urls.
Call this tool with input_url = get_url plus any extras you want.
Download the result from the returned download_url (default 1h TTL, max 24h).

Supported formats

Video output	Audio output
mp4, webm, gif, mkv, avi	mp3, aac, wav, flac, ogg

Input format is auto-detected. Subtitle files: SRT, VTT, or ASS.

Options

Core (single-input transcode)

Name	Type	Default	Description
`input_url`	string	required	Pre-signed GET URL of the source file.
`output_format`	string	required	Target format. Audio formats strip video automatically.
`output_filename`	string	`output.<ext>`	Filename in the download URL.
`expires_in`	integer	`3600`	Download URL TTL in seconds (60–86400).
`video_crf`	integer	-	Video quality (CRF 0–51). 18 = visually lossless, 23 = default. Lower is better.
`audio_bitrate`	string	-	Audio bitrate, e.g. `128k`, `192k`, `320k`.
`resolution`	string	-	Output resolution, e.g. `1280x720`. Use `-2` to auto-scale, e.g. `1280:-2`.
`start_time`	number	-	Clip start in seconds. Single-input mode only.
`end_time`	number	-	Clip end in seconds. Single-input mode only.

Multi-clip and effects (optional)

Name	Type	Description
`concat`	array of URLs	Additional clip URLs to concatenate after `input_url`. Output = primary + concat clips in order. Up to 20 extras.
`overlay`	object	Overlay another video/image on top of the primary input. Fields: `url`, `x`, `y`, optional `start_time`, `end_time`.
`subtitles_url`	string	Pre-signed GET URL of an SRT/VTT/ASS subtitle file to burn into the video.
`text`	object	Burn a single text label into the video, optionally with a background box and time window. See fields below.

Overlay object

Field	Type	Default	Description
`url`	string	required	Pre-signed GET URL of the overlay video or image.
`x`	integer	0	Horizontal position of overlay top-left (px).
`y`	integer	0	Vertical position of overlay top-left (px).
`start_time`	number	-	Time in seconds when overlay appears. Omit for full duration.
`end_time`	number	-	Time in seconds when overlay disappears.

Text object

Draws a single text label, optionally inside a coloured box, optionally only between start_time and end_time.

Field	Type	Default	Description
`text`	string	required	Text to burn into the video.
`x`	integer	10	Horizontal position (px).
`y`	integer	10	Vertical position (px).
`font_size`	integer	32	Font size in pixels (6–400).
`color`	string	`white`	Text color (CSS name or hex).
`background_color`	string	-	If set, a coloured box is drawn behind the text.
`background_opacity`	number	0.5	Box opacity 0.0–1.0.
`padding`	integer	8	Padding (px) around the text inside the box.
`start_time`	number	-	Time in seconds when text appears. Omit for full duration.
`end_time`	number	-	Time in seconds when text disappears.

Output

{
  "download_url": "https://…/output/…/output.mp4",
  "key": "output/…",
  "size_bytes": 4821903,
  "mime": "video/mp4",
  "input_bytes": 52428800,
  "duration_seconds": 62.4,
  "expires_at": "2026-05-15T13:00:00Z"
}

Pricing

1 credit per second of output duration. A 60-second video costs 60 credits. Concatenating five 30-second clips into one 150-second video costs 150 credits. Adding an overlay or subtitles doesn't change the price, output duration is what counts.

Max input size: 500 MB per file. Max encode time: 10 minutes.

GIF output

GIF output is automatically downsampled to 10 fps for manageable file sizes. For a custom resolution, set the resolution field.