Video Segment Extraction

Execution Priority

Prioritize calling this skill's scripts/run.py entrypoint first. Only write custom code after the script result is unsuitable or the script cannot cover the requested task.

Input Resolution

If event timing is missing but the user provided a video analysis request, do not write custom clipping code. First call upstream event skills, especially event-rule-engine/scripts/run.py, to produce an event with event_time or event_elapsed_seconds; then call this skill with raw_segment_uri, event_id, and the pre/post window.

Atomic CLI

Run this skill directly with its own script. The script does not call other skill scripts and does not depend on shared src, tools, or registry modules.

python video-segment-extraction/scripts/run.py --event-id <event_id> --raw-segment-uri <video.mp4> --event-time <iso_time> --event-elapsed-seconds <seconds> --config <config.json> --output <clip.json>

Parameters: --event-id, --raw-segment-uri, --event-time, --start-time, --event-elapsed-seconds, --pre-seconds, --post-seconds, --output-dir, --config, --output.

Workflow

1. Read event id, source segment, event time, and pre/post window.
2. Locate the source media range.
3. Clip or synthesize a short segment in a playable format.
4. Store and return the clip URI and time range.

Available Implementation

This skill is implemented as an atomic standalone script in its own scripts/run.py. The script contains the executable logic for this skill and must not import shared src, tools, or registry modules.

Inputs

Required:

• event_id
• event_time or start_time

Common fields:

• camera_id
• raw_segment_uri
• event_elapsed_seconds
• pre_seconds
• post_seconds

Outputs

Returns event_id, clip_uri, start_time, and end_time.

Failure Modes

• MISSING_EVENT
• FFMPEG_CLIP_FAILED

Constraints

Do not decide whether an event occurred. Do not modify event confidence.