Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    shipreel

    1

    Generate narrated, animated PR walkthrough videos from diffs and UI recordings.

    Free

    Security scanned
    shipreel

    shipreel

    Example session with this skill installed

    Make a shipreel for PR #42 emphasizing the new auth flow.

    • Read your context and instructions
    • Compiled the shipreel

    I have analyzed the diff for PR #42 (JWT Authentication Refactor). I've planned a 145-second walkthrough across 5 scenes: Intro, Token Validation, Middleware Logic, Login UI, and Reviewer Checklist.

    I recorded a UI clip of the new login flow and generated animated diagrams for the middleware chain. The build is complete.

    Walkthrough Video Generated

    File: ./.shipreel/pr-42/out/pr-42-auth-refactor.mp4
    Duration: 02:24
    Size: 8.4 MB

    Chapters

    • 0:00 Intro: Replacing session-based auth with JWT
    • 0:35 How the token validation logic changed
    • 1:10 The new AuthMiddleware implementation
    • 1:55 UI Demo: Login and Redirect flow
    • 2:15 Reviewer focus: Verify token expiration handling

    Next steps

    1. Run open ./.shipreel/pr-42/out/pr-42-auth-refactor.mp4 to preview the video.
    2. I can add a "Walkthrough video" section to your PR description. Should I proceed?
    3. Once I add the section, you will need to manually drag the MP4 file into the GitHub PR editor.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Reviewing a massive diff is a high-cognitive load task that often leads to shallow approvals. Shipreel reduces this friction by generating 150-second narrated walkthrough videos directly from your pull request data, combining animated diagrams, code snippets, and UI recordings.

    What it does

    • Automated narration converts PR descriptions and diffs into a concise, spoken script using local system voices.
    • Animated visuals generates SVG-based diagrams, code panels, and labels to explain backend logic and architectural changes.
    • UI recording orchestrates headless browser sessions to capture real application flows for frontend changes.
    • Scene orchestration builds a multi-scene MP4 with crossfades, ambient music, and synchronized captions.
    • GitHub optimization targets specific file sizes to ensure the output fits within the 10MB limit for direct PR attachments.

    How it works

    1. Analyze the PR by fetching the diff and metadata via the GitHub CLI to identify the core changes.
    2. Draft the story by selecting 4 to 7 key scenes that focus on the "why" and "how" rather than a file-by-file tour.
    3. Generate the scenes using a JavaScript-based engine that produces SVG animations and records optional UI clips.
    4. Render the video locally using FFmpeg to combine audio, video, and transitions into a final MP4 file.

    Frameworks & tools

    Built for Node.js environments utilizing the GitHub CLI (gh), FFmpeg, and Chrome. The engine uses a custom library for SVG generation and Puppeteer for headless UI recording.

    Why this beats prompting it yourself

    Writing a script, drawing diagrams, and recording your screen manually takes hours per PR. This skill automates the entire production pipeline, ensuring consistent timing, professional pacing, and local execution with no recurring API costs for video generation.

    Use cases

    • Explain complex refactors where the diff alone doesn't show the new architectural flow.
    • Showcase new UI features to stakeholders before they pull the branch locally.
    • Onboard new team members to a codebase by providing a library of video-documented changes.
    • Highlight specific areas in a large PR that require extra scrutiny from senior reviewers.

    Known limitations

    Requires local installations of Chrome and FFmpeg. Narration quality depends on the host system's text-to-speech engine unless configured for external services. Video uploads to GitHub must be done manually via the web interface.

    How to use

    Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scanned

    Verified clean 13 days ago

    • Free to download with an account

    Listed13 days ago

    What's inside

    Frequently Asked Questions