Evals / video / See inside a video

Sample frames

Save one frame every 2 seconds of bunny.mp4 as images in a folder called frames.

The task

The reference answer shows the job can be done with deft. It is not an agent run.

deft video frames bunny.mp4 --interval 2 -o frames/

Input

bunny.mp4 946.5 KB · h264 1280x720, 10.0s

Reference output

frames/bunny-0001.jpg 224.9 KB · 1280x720
frames/bunny-0001.jpg
frames/bunny-0002.jpg 234.1 KB · 1280x720
frames/bunny-0002.jpg
frames/bunny-0003.jpg 234.8 KB · 1280x720
frames/bunny-0003.jpg
frames/bunny-0004.jpg 235.9 KB · 1280x720
frames/bunny-0004.jpg
frames/bunny-0005.jpg 239.4 KB · 1280x720
frames/bunny-0005.jpg
How a result is judged (1 check)
  • count_files: glob = "frames/*", min = 5, max = 6

Runs

CodexWithout deftPassed2026-11-07

Time
12s
New tokens
3.3k
Total input
40.1k
Output tokens
0.3k
Commands
2
Failed
0

Trace

  1. I’ll check the video and available media tools, then save frames at two-second intervals in `frames/`.

  2. pwd && rg --files -g 'bunny.mp4' -g 'AGENTS.md' -g '"'!*node_modules*'"' | head -80 && command -v ffmpeg || true && command -v python3 || true
    output
    bunny.mp4
    /opt/homebrew/bin/ffmpeg
    /usr/local/bin/python3
  3. mkdir -p frames && ffmpeg -hide_banner -loglevel error -i bunny.mp4 -vf "fps=1/2" frames/frame_%04d.png && find frames -maxdepth 1 -type f | sort | wc -l
    output
    5
  4. Saved 5 frames from `bunny.mp4` to the [frames folder](frames), one every 2 seconds.

Output

frames/frame_0001.png 1.9 MB · 1280x720
frames/frame_0001.png
frames/frame_0002.png 2.0 MB · 1280x720
frames/frame_0002.png
frames/frame_0003.png 2.0 MB · 1280x720
frames/frame_0003.png
frames/frame_0004.png 2.0 MB · 1280x720
frames/frame_0004.png
frames/frame_0005.png 2.0 MB · 1280x720
frames/frame_0005.png

CodexWith deft, skill installedPassedRead the skillUsed deft2026-11-07

Time
11s
New tokens
4.3k
Total input
56.2k
Output tokens
0.2k
Commands
3
Failed
0

Trace

  1. I’ll use the `deft` skill for video frame extraction and save the images under `./frames`.

  2. cat .agents/skills/deft/SKILL.md
    output
    ---
    name: deft
    description: "Use deft instead of writing a script for file jobs: PDF merge/split/compress/watermark, image resize/convert/compress, video and audio trim/convert/compress, spreadsheets, Markdown to PDF/HTML, and YouTube downloads and subtitles."
    ---
    
    # deft
    
    One local command for file jobs. Prefer it over writing Python or shell scripts.
    
    Syntax: `deft <group> <command> [inputs] [-o output] [--options]`
    
    Groups:
    - `pdf`: Merge, split, compress, encrypt, forms, OCR, compare, convert
    - `image`: Resize, crop, convert, compress, filters, montage, OCR
    - `video`: Trim, crop, speed, convert, compress, GIF, frames (ffmpeg)
    - `audio`: Convert, trim, merge, fade, speed, tag, normalize (ffmpeg)
    - `doc`: Word/ODT/RTF: to text, Markdown, HTML, PDF; find and replace
    - `slides`: PowerPoint: extract text, Markdown outline, PDF handout
    - `data`: JSON/YAML/TOML/XML: convert, query with jq syntax, merge, diff
    - `sheet`: Excel/CSV/ODS tables: filter, sort, join, pivot, convert
    - `md`: Markdown to PDF, HTML, Word, EPUB; table of contents
    - `html`: HTML to Markdown, text, Word, PDF; screenshots
    - `epub`: E-books: info, to text, Markdown and PDF
    - `email`: EML files: headers, text, attachments
    - `contact`: VCF address books: to and from CSV
    - `calendar`: ICS calendars: to and from CSV
    - `gps`: GPX and KML tracks: convert, distance and time
    - `youtube`: Info, download, audio, subtitles, thumbnail (yt-dlp)
    - `file`: Any file: hash, duplicates, batch rename, base64, encoding
    - `qr`: Cre
    … 728 more characters
  3. deft video frames --help
    output
    deft video frames - Save frames as images: --fps 1 or --at 0:05,0:10
    
    Usage: deft video frames <video>... [-o <dir>] [options]
    In: video (each handled separately)   Out: png|jpg files in <name>-frames/
    
    Options:
      --fps <number>               Frames per second
      --interval <time>            Time between frames, such as 2 or 0:30
      --from <time>                Start time for --fps or --interval (default 0)
      --at <times>                 Exact times
      --to <png|jpg>               Image format (default jpg)
    Common: -o --force --dry-run --json -j   (deft help common)
    
    Examples:
      deft video frames clip.mp4 --at 0:01 -o shots/
      deft video frames clip.mp4 --fps 1 --to png -o frames/
      deft video frames clip.mp4 --from 1 --interval 2 -o frames/
    
    Result fields: files
    Needs: ffmpeg, ffprobe
  4. deft video frames bunny.mp4 --interval 2 -o frames
    output
    ok video.frames bunny.mp4 -> frames/  5 frames  0.17s
  5. Saved 5 frames from `bunny.mp4` to [frames](frames).

Output

frames/bunny-0001.jpg 100.8 KB · 1280x720
frames/bunny-0001.jpg
frames/bunny-0002.jpg 129.8 KB · 1280x720
frames/bunny-0002.jpg
frames/bunny-0003.jpg 237.0 KB · 1280x720
frames/bunny-0003.jpg
frames/bunny-0004.jpg 236.7 KB · 1280x720
frames/bunny-0004.jpg
frames/bunny-0005.jpg 189.5 KB · 1280x720
frames/bunny-0005.jpg

CodexWith deft, skill as AGENTS.mdPassed2026-11-05

Time
13s
New tokens
5.9k
Total input
61.7k
Output tokens
0.4k
Commands
4
Failed
0

Trace

  1. I’ll use the local `deft` video frame extraction command, first checking its options and the input file, then save the frames under `./frames`.

  2. deft video
    output
    deft video - commands (input -> output)
      info           video -> report      Duration, size, frame rate, codecs
      compress       video -> mp4         Smaller file: --level, --target-size, --max-height
      convert        video -> video       Change container and codec: --to mp4|webm|mov|mkv
      download       url -> video         Save a video: --resolution 720p, --to mp4|webm|mkv [net]
      extract-audio  video -> audio       Save the soundtrack: --to mp3|m4a|wav|flac|ogg
      frames         video -> png|jpg...  Save frames as images: --fps 1 or --at 0:05,0:10
      merge          video... -> mp4      Join videos in order (sizes are matched to the first)
      mute           video -> video       Remove the audio track
      resize         video -> video       Scale: --size 1280x, x720, 50%
      rotate         video -> video       Turn clockwise: --angle 90
      split          video -> video       Cut out a part: --from 0:10 --to-time 0:25
      to-gif         video -> gif         Animated GIF: --fps, --size, --from, --to-time
    More: deft video <command> --help
  3. deft video frames --help
    output
    deft video frames - Save frames as images: --fps 1 or --at 0:05,0:10
    
    Usage: deft video frames <video>... [-o <dir>] [options]
    In: video (each handled separately)   Out: png|jpg files in <name>-frames/
    
    Options:
      --fps <number>               Frames per second
      --at <times>                 Exact times
      --to <png|jpg>               Image format (default jpg)
    Common: -o --force --dry-run --json -j   (deft help common)
    
    Examples:
      deft video frames clip.mp4 --at 0:01 -o shots/
      deft video frames clip.mp4 --fps 1 --to png -o frames/
    
    Result fields: files
    Needs: ffmpeg, ffprobe
  4. ls -l bunny.mp4
    output
    -rw-r--r--@ 1 user  staff  969201 Nov  4 19:32 bunny.mp4
  5. deft video frames bunny.mp4 --fps 0.5 --to jpg -o frames
    output
    ok video.frames bunny.mp4 -> frames/  5 frames  0.17s
  6. Saved 5 frames, one every 2 seconds, in [frames](<frames>).

Output

Its files were not kept in this run. The last line of the trace is its own report.

Claude Opus 5.5Without deftPassed2026-11-07

Time
11s
New tokens
8.5k
Total input
41.9k
Output tokens
0.4k
Commands
0
Failed
0

Trace

    Output

    frames/frame_0001.png 1.9 MB · 1280x720
    frames/frame_0001.png
    frames/frame_0002.png 2.0 MB · 1280x720
    frames/frame_0002.png
    frames/frame_0003.png 2.0 MB · 1280x720
    frames/frame_0003.png
    frames/frame_0004.png 2.0 MB · 1280x720
    frames/frame_0004.png
    frames/frame_0005.png 2.0 MB · 1280x720
    frames/frame_0005.png

    Claude Opus 5.5With deft, skill installedPassedRead the skillUsed deft2026-11-07

    Time
    9s
    New tokens
    9.9k
    Total input
    59.4k
    Output tokens
    0.4k
    Commands
    0
    Failed
    0

    Trace

      Output

      ---
      name: deft
      description: "Use deft first for any simple, direct operation on local files, whatever the type: convert, read, extract, edit, compress, merge, split, compare or inspect. Try it before writing a script or one-liner. If unsure, run deft find; if it says no command does this, use another tool. Not for programming work: to read, write or change source code, skip it."
      ---
      
      # deft
      
      One local command for file jobs. Prefer it over writing Python or shell scripts.
      
      Syntax: `deft <group> <command> [inputs] [-o output] [--options]`
      
      Groups:
      - `pdf`: Merge, split, compress, encrypt, forms, OCR, compare, convert
      - `image`: Resize, crop, convert, compress, adjust, montage, OCR
      - `video`: Trim, merge, convert, compress, subtitles, GIF, frames (ffmpeg)
      - `audio`: Convert, trim, normalize, denoise, cut silence (ffmpeg)
      - `doc`: Word/ODT/RTF: to text, Markdown, HTML, PDF; find and replace
      - `slides`: PowerPoint: extract text, Markdown outline
      - `sheet`: Excel/CSV/ODS tables: SQL query, pivot, merge, split, convert
      - `json`: JSON: convert, format, merge, diff, validate
      - `toml`: TOML: convert, format, merge, diff, validate
      - `xml`: XML: convert, format, merge, diff, validate
      - `yaml`: YAML: convert, format, merge, diff, validate
      - `md`: Markdown to PDF, HTML, Word, EPUB; table of contents
      - `html`: HTML to Markdown, text, Word and PDF
      - `epub`: E-books: info, to text, Markdown and PDF
      - `gps`: GPX and KML tracks: convert, distance and time
      - `qr`: Create QR codes and barcodes; read QR codes
      
      How to use it:
      1. Run `deft <group>` to list commands, or `deft find "<what you want>"` to search. If it says `no command does this`, use another tool.
      2. Run `deft <group> <command> --help` for options and examples.
      3. `deft convert <in> -o <out.ext>` converts between formats by extension.
      4. Results are one line starting with `ok`, `partial` or `error`; add `--json` to parse them. Errors include a `hint:` with the fix.
      
      Inputs are never overwritten; pass `--force` or `--in-place` to allow it. Globs and directories work as inputs; several inputs run in parallel.
      
      If `deft` is not installed, ask the user, then follow https://github.com/firstprayer/deft/blob/main/docs/INSTALL.md
      
      frames/bunny-0001.jpg 224.9 KB · 1280x720
      frames/bunny-0001.jpg
      frames/bunny-0002.jpg 234.1 KB · 1280x720
      frames/bunny-0002.jpg
      frames/bunny-0003.jpg 234.8 KB · 1280x720
      frames/bunny-0003.jpg
      frames/bunny-0004.jpg 235.9 KB · 1280x720
      frames/bunny-0004.jpg
      frames/bunny-0005.jpg 239.4 KB · 1280x720
      frames/bunny-0005.jpg

      Claude CodeWithout deftNot run yet

      Claude CodeWith deftNot run yet