VideoDB Python Skill

A single API-first video stack for agents. Ingest anything, process server-side, and ship playable streams without FFmpeg glue.

Core capabilities

• Ingest + Transcode — Accept any format, change codec/bitrate/FPS/resolution, output a playable stream (CDN + hosting)
• Scene-level Search Engine — Build a searchable index scene by scene, find exact moments and auto-create clips, manage 1000s of hours of footage
• Generate + Compose — Generate image/audio/video/text assets, overlay text/images/branding/motion captions, dub videos, translate captions
• Real-time RTSP — Connect live streams, define events and alerts, ideal for security cams and monitoring
• Desktop Perception — Capture screen/mic/system audio, stream desktop live, define alerts and triggers, store episodic memory and search sessions

Try it now

• “Ingest this file and give me a playable web stream link”
• “Generate subtitles, burn them in, and add light background music”
• “Index this folder and find every scene with people”
• “Connect this RTSP URL and alert when a person enters the zone”
• “Start recording and give me an actionable summary when it ends”

Running Python code

scripts/videodb_env.py handles environment loading and SDK validation. Call load_vdb_env() before any VideoDB code — it checks the SDK is installed, loads VIDEO_DB_API_KEY from environment variable → ./.env → ~/.videodb/.env, and exits with clear error messages if anything is missing. All scripts in scripts/ already call this at the top.

Inline usage:

python -c "from scripts.videodb_env import load_vdb_env; load_vdb_env(); print('OK')"

Usage in a script file:

from scripts.videodb_env import load_vdb_env
load_vdb_env()

import videodb
conn = videodb.connect()

videodb.connect() reads VIDEO_DB_API_KEY from the environment automatically.

Do NOT write a script file when a short inline command works.

When writing inline Python (python -c "..."), always use properly formatted code — use semicolons to separate statements and keep it readable. For anything longer than ~3 statements, use a heredoc instead:

python << 'EOF'
from scripts.videodb_env import load_vdb_env
load_vdb_env()

import videodb
conn = videodb.connect()
coll = conn.get_collection()
print(f"Videos: {len(coll.get_videos())}")
EOF

Setup

When the user asks to “setup videodb” or similar:

1. Install SDK

pip install "videodb[capture]" python-dotenv

If videodb[capture] fails on Linux, install without the capture extra:

pip install videodb python-dotenv

2. Verify environment

Run load_vdb_env() to verify the SDK and API key:

python -c "from scripts.videodb_env import load_vdb_env; load_vdb_env(); print('OK')"

If it prints OK, setup is done.

If it fails with VIDEO_DB_API_KEY not found, ask the user to:

1. Get a free API key at https://console.videodb.io (50 free uploads, no credit card)
2. Set it up using either of these methods:
   • Set it as an environment variable: export VIDEO_DB_API_KEY=your-key
   • Or save it to ~/.videodb/.env as VIDEO_DB_API_KEY=your-key

Then re-run to confirm.

Do NOT read, write, or handle the API key yourself. Always let the user set it.

Quick Reference

Upload media

# URL
video = coll.upload(url="https://example.com/video.mp4")

# YouTube
video = coll.upload(url="https://www.youtube.com/watch?v=VIDEO_ID")

# Local file
video = coll.upload(file_path="/path/to/video.mp4")

Transcript + subtitle

# force=True skips the error if the video is already indexed
video.index_spoken_words(force=True)
text = video.get_transcript_text()
stream_url = video.add_subtitle()

Search inside videos

from videodb.exceptions import InvalidRequestError

video.index_spoken_words(force=True)

# search() raises InvalidRequestError when no results are found.
# Always wrap in try/except and treat "No results found" as empty.
try:
    results = video.search("product demo")
    shots = results.get_shots()
    stream_url = results.compile()
except InvalidRequestError as e:
    if "No results found" in str(e):
        shots = []
    else:
        raise

Scene search

import re
from videodb import SearchType, IndexType, SceneExtractionType
from videodb.exceptions import InvalidRequestError

# index_scenes() has no force parameter — it raises an error if a scene
# index already exists. Extract the existing index ID from the error.
try:
    scene_index_id = video.index_scenes(
        extraction_type=SceneExtractionType.shot_based,
        prompt="Describe the visual content in this scene.",
    )
except Exception as e:
    match = re.search(r"id\s+([a-f0-9]+)", str(e))
    if match:
        scene_index_id = match.group(1)
    else:
        raise

# Use score_threshold to filter low-relevance noise (recommended: 0.3+)
try:
    results = video.search(
        query="person writing on a whiteboard",
        search_type=SearchType.semantic,
        index_type=IndexType.scene,
        scene_index_id=scene_index_id,
        score_threshold=0.3,
    )
    shots = results.get_shots()
    stream_url = results.compile()
except InvalidRequestError as e:
    if "No results found" in str(e):
        shots = []
    else:
        raise

Timeline editing

Important: Always validate timestamps before building a timeline:

• start must be >= 0 (negative values are silently accepted but produce broken output)
• start must be < end
• end must be <= video.length

from videodb.timeline import Timeline
from videodb.asset import VideoAsset, TextAsset, TextStyle

timeline = Timeline(conn)
timeline.add_inline(VideoAsset(asset_id=video.id, start=10, end=30))
timeline.add_overlay(0, TextAsset(text="The End", duration=3, style=TextStyle(fontsize=36)))
stream_url = timeline.generate_stream()

Transcode video (resolution / quality change)

from videodb import TranscodeMode, VideoConfig, AudioConfig

# Change resolution, quality, or aspect ratio server-side
job_id = conn.transcode(
    source="https://example.com/video.mp4",
    callback_url="https://example.com/webhook",
    mode=TranscodeMode.economy,
    video_config=VideoConfig(resolution=720, quality=23, aspect_ratio="16:9"),
    audio_config=AudioConfig(mute=False),
)

Reframe aspect ratio (for social platforms)

Warning: reframe() is a slow server-side operation. For long videos it can take several minutes and may time out. Best practices:

• Always limit to a short segment using start/end when possible
• For full-length videos, use callback_url for async processing
• Trim the video on a Timeline first, then reframe the shorter result

from videodb import ReframeMode

# Always prefer reframing a short segment:
reframed = video.reframe(start=0, end=60, target="vertical", mode=ReframeMode.smart)

# Async reframe for full-length videos (returns None, result via webhook):
video.reframe(target="vertical", callback_url="https://example.com/webhook")

# Presets: "vertical" (9:16), "square" (1:1), "landscape" (16:9)
reframed = video.reframe(start=0, end=60, target="square")

# Custom dimensions
reframed = video.reframe(start=0, end=60, target={"width": 1280, "height": 720})

Generative media

image = coll.generate_image(
    prompt="a sunset over mountains",
    aspect_ratio="16:9",
)

Error handling

from videodb.exceptions import AuthenticationError, InvalidRequestError

try:
    conn = videodb.connect()
except AuthenticationError:
    print("Check your VIDEO_DB_API_KEY")

try:
    video = coll.upload(url="https://example.com/video.mp4")
except InvalidRequestError as e:
    print(f"Upload failed: {e}")

Common pitfalls

Additional docs

Reference documentation is in the reference/ directory adjacent to this SKILL.md file. Use the Glob tool to locate it if needed.

• reference/api-reference.md – Complete VideoDB Python SDK API reference
• reference/search.md – In-depth guide to video search (spoken word and scene-based)
• reference/editor.md – Timeline editing, assets, and composition
• reference/generative.md – AI-powered media generation (images, video, audio)
• reference/rtstream.md – Real-time streaming capabilities
• reference/capture.md – Screen and audio capture
• reference/use-cases.md – Common video processing patterns and examples

Screen Recording (Desktop Capture)

Use scripts/capture_bg.py for screen and audio recording with AI transcription and visual indexing.

Start Recording

Run in background mode:

python scripts/capture_bg.py start &

The recording will capture:

• Screen video
• Microphone audio
• System audio
• Real-time AI transcription
• Visual scene descriptions

Check Status

cat /tmp/videodb_capture_state.json

Stop Recording

Create the stop file to signal recording to finish:

touch /tmp/videodb_capture_stop

Or run:

python scripts/capture_bg.py stop

Get Shareable Link

After stopping, the state file contains the video URL:

cat /tmp/videodb_capture_state.json

Returns JSON with player_url for sharing.

Utility scripts

Ready-to-run scripts are in the scripts/ directory adjacent to this SKILL.md file. Read and execute them directly instead of rewriting the logic.

• scripts/videodb_env.py – Environment loading and SDK validation (called by all scripts)
• scripts/capture_bg.py – Screen recording with AI (recommended for capture)
• scripts/capture.py – Interactive screen capture (requires terminal input)
• scripts/batch_upload.py – Bulk upload from a URL list or directory
• scripts/search_and_compile.py – Search inside a video and compile matching clips into a stream
• scripts/extract_clips.py – Extract clips by timestamp ranges
• scripts/backend.py – Capture backend server (WebSocket polling, no tunnel required)
• scripts/client.py – Capture client (connects to backend)
• scripts/check_connection.py – Verify API key and connection

Do not use ffmpeg, moviepy, or local encoding tools when VideoDB supports the operation. The following are all handled server-side by VideoDB — trimming, combining clips, overlaying audio or music, adding subtitles, text/image overlays, transcoding, resolution changes, aspect-ratio conversion, resizing for platform requirements, transcription, and media generation. Only fall back to local tools for operations listed under Limitations in reference/editor.md (transitions, speed changes, crop/zoom, colour grading, volume mixing).

When to use what