video-summarizer
Skill21 starsChanged 7 months agoMITView on GitHub
Analyzes local MP4 video files using the Gemini API and generates structured Markdown summaries. Supports custom prompts to extract specific information instead of the default summary. Use when the user wants to summarize, analyze, or extract key points from video files in bulk. Triggers on requests like "summarize my videos", "create markdown summaries for the videos", "analyze videos with Gemini", "generate video summaries", or "extract X from my videos". Requires GEMINI_API_KEY environment variable set in the system.
- Reads credentials
- Installs packages
2 other projects use a version of this.
Adopted by
Most recently taken up by
- LingyiChen-AI/skillstheir own version, first seen Nov 5, 2026
- Jiguang-zero/video-summarizer-skilltheir own version, first seen Aug 19, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- Jiguang-zero/video-summarizer-skill★ 1same name
- LingyiChen-AI/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository22 other files it gives its agents.
- 2d-pixel-assetSkill
- anthropic-skilljar-extractorSkill
- audio-to-midiSkill
- auto-researchSkill
- gemini-image-genSkill
- gemini-web-media-analysisSkill
- google-ai-studio-ttsSkill
- hermes-tweetSkill
- link-downloadSkill
- llama-cppSkill
- meta-muse-video-analysisSkill
- midi-synthSkill
- opencodeSkill
- seedanceSkill
- tech-product-landingSkill
- unreal-best-practicesSkill
- unreal-blueprint-codegenSkill
- unreal-gasSkill
- unreal-pcg-pythonSkill
- unreal-thirdpartySkill
- video-downloadSkill
- video-fetch-and-summarizeSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Check if the user provided a custom prompt or specific extraction instructions.
- Distinctive: line 57
When the user invokes this skill:
- Very specific: line 59
If the user specifies a custom prompt or extraction goal (e.g., "extract all tools mentioned", "summarize focusing on pricing", "list …
- Very specific: line 61
If the user provides a prompt file path, pass it via --prompt-file path/to/file.txt.
- Very specific: line 62
If the user gives no specific instructions, run without --prompt to use the default format.
- Distinctive: line 63
If the user specifies a directory, pass it as the first positional argument.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Video Summarizer
Analyzes MP4 files in a directory via the Gemini API and writes a .md summary file next to each video.
Prerequisites
GEMINI_API_KEYenv var must be setgoogle-genaiinstalled:py -m pip install google-genai- Python available as
py(Windows) orpython3
Workflow
1. Determine the prompt
Check if the user provided a custom prompt or specific extraction instructions.
- No custom prompt -> use the default (structured summary with Title, Summary, Key Points, Takeaways)
- User provides custom instructions -> pass them via
--promptor--prompt-file
A custom prompt completely replaces the default. Examples of custom prompts:
- "Extract all mentioned tools, their prices, and a one-line description for each"
- "List every action item and deadline mentioned in this video"
- "Create a transcript outline with timestamps"
2. Run the script
# Default summary — all videos in current directory (recursive)
py scripts/summarize_videos.py
# Default summary — specific directory
py scripts/summarize_videos.py "C:/path/to/videos"
# Custom prompt (inline) — replaces default summary format
py scripts/summarize_videos.py --prompt "Extract all product names and prices mentioned"
# Custom prompt (from file) — replaces default summary format
py scripts/summarize_videos.py --prompt-file my-prompt.txt
# Both directory and custom prompt
py scripts/summarize_videos.py "C:/path/to/videos" --prompt "List all key decisions made"3. How Claude should handle user arguments
When the user invokes this skill:
- If the user specifies a custom prompt or extraction goal (e.g., "extract all tools mentioned",
"summarize focusing on pricing", "list action items"), pass it via
--prompt "...". - If the user provides a prompt file path, pass it via
--prompt-file path/to/file.txt. - If the user gives no specific instructions, run without
--promptto use the default format. - If the user specifies a directory, pass it as the first positional argument.
4. Default output format (when no custom prompt is given)
# [Video Title]
## Summary
[2-3 sentence description]
## Key Points
- point 1
- point 2
- ...
## Core Content & Takeaways
[Central message and key insight]
---
*Auto-generated with Gemini API*Notes
- Language of the summary matches the language of the video
- Files are auto-deleted from Gemini storage after processing (48h limit)
- Already-summarized videos are skipped on re-runs
- Script path:
scripts/summarize_videos.py - Model:
gemini-3-flash-preview
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

