Design & media

Google Video URL Analysis

Try it

Send any public video URL to Gemini for text-based analysis — summaries, Q&A, timestamps, and moderation.

What it does

Analyzes publicly accessible videos by sending their URL directly to a Google Gemini model. The analyzer accepts a public video URL and a task-specific prompt, then returns text output. Supports multiple analysis types: summaries, question-answering about content, timeline extraction, and content moderation. Run via a Python script with an API key and a concrete prompt specifying the desired output format.

When to use it

  • Summarizing a video into bullet points and timestamps
  • Asking natural language questions about a video's content
  • Extracting a structured timeline of events from a recording
  • Checking video content against moderation guidelines

The skill document

Google Video Url Analysis

Use scripts/gemini_video_url_analyzer.py for deterministic Gemini calls instead of ad-hoc API snippets.

Workflow

  1. Set an API key:
    • MAVERICK_GEMINI_API_KEY (preferred), or
    • MAVERICK_GOOGLE_API_KEY
  2. Run the analyzer with a public video URL supported by Gemini as direct URI input.
  3. Provide a task-specific prompt (summary, timeline extraction, Q&A, moderation, etc.).
  4. Inspect the text output.

Command Guide

  • Basic analysis:
    • python scripts/gemini_video_url_analyzer.py --video-url "https://www.youtube.com/watch?v=..." --prompt "Summarize in 5 bullets and list important timestamps."
  • Override model:
    • python scripts/gemini_video_url_analyzer.py --video-url "" --model "gemini-2.5-pro" --prompt "Produce a scene-by-scene report."

Operational Rules

  • Require a publicly accessible URL.
  • Prefer concise, explicit prompts with concrete output constraints.

Questions people ask

What video URLs are supported?
The tool accepts any public video URL that Gemini supports as direct URI input. The document shows YouTube URLs as examples.
What API key do I need?
Set the `MAVERICK_GEMINI_API_KEY` environment variable. The `MAVERICK_GOOGLE_API_KEY` is also accepted as a fallback.
What does the output look like?
The analyzer returns plain text output. Prompt engineering determines the format — for example, requesting '5 bullets and timestamps' or 'scene-by-scene report' shapes the output structure.

Related skills

Join video meetings as a voice bot, visual avatar, or avatar with live screen sharing.

by johnpatternai22 installs8 stars

Escape the scarcity trap — diagnose bandwidth consumption and design protected slack to restore strategic capacity.

by deciqai1 installs2 stars

Tests whether you genuinely understand something or just recognize it — exposes the gaps in your mental model.

by deciqai1 installs2 stars

Make irreversible life decisions by projecting to 80 and naming which regret you'd rather live with.

by deciqai1 installs2 stars

Turn China 3C launch inputs into executable routes, messaging, channel actions, risk checks, and review decisions.

by killsnake0126 installs112 stars

Diagnose which mental domain is holding you back before choosing a cognitive intervention.

by deciqai1 installs3 stars

More from maverick

Browse all skills

Read permitted HubSpot data through a dynamically discovered, read-only MCP tool catalog.

by maverick28 installs

Read and write PandaDoc workspace data through PandaDoc’s hosted MCP server.

by maverick15 installs

Research X posts and users, search context, and execute individually confirmed X actions.

by maverick28 installs

Read and update Linear workspace data through Linear’s hosted MCP server.

by maverick18 installs

Read and manage WordPress.com content through its hosted MCP server and live tool catalog.

by maverick28 installs

Discover and invoke Canva’s current MCP tools while respecting account permissions and confirmation boundaries.

by maverick30 installs