More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    History Reels Generator

    1

    History and educational content often struggle with retention on social media or rely on low-quality AI-generated imagery.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    History Reels Generator

    History Reels Generator

    Example session with this skill installed

    Make a Reel explaining Napoleon's retreat from Russia in 1812.

    • Read your context and instructions
    • Compiled the history reels generator

    Phase 1: script for approval

    Real output from a published Reel, translated from Portuguese (the skill writes in the user's language). Assumptions: lay audience, 1080×1920 Reel, 30 fps, sober narration with free edge-tts voice. Thesis: the cold was only part of the disaster; hunger, disease, exhaustion and Russian attacks destroyed the army, and Minard turned it into a drawing.

    TimeNarrationImage / mapDrawingLabel
    0:00Moscow, 19 October 1812. Napoleon has taken the city, but it is mostly burned and without food. After five weeks, his army leaves.Moscow in flames, Albrecht Adam, 1841 (PD)—19 OCT 1812, MOSCOW
    0:12He tries to go back through the south. At Maloyaroslavets, the Russians block the way.Map printed in 1812, BnF (PD)Red line Moscow → Maloyaroslavets, battle pinROAD BLOCKED
    0:18So he retreats north, towards Smolensk, on a route he wanted to avoid.Same mapLine to Smolensk over the printed symbolsLATE OCT 1812
    0:24In November come snow and cold. But hunger, disease and exhaustion were already destroying the army.Pryanishnikov, 1874 (PD)—HUNGER AND DISEASE
    0:35The Berezina river blocks the way, with the Russians close.1812 mapLine to Borisov, blue circleBEREZINA RIVER
    0:49On 29 November the bridges are destroyed. Almost 10,000 people are left on the other side.Anonymous watercolour, Musée de l'Armée (PD)—ALMOST 10,000
    1:08In 1869, engineer Charles Minard drew this tragedy. By his numbers, 422,000 went in, 100,000 reached Moscow, about 10,000 came back.Minard's chart, 1869, BnF (PD)Labels 422K and 100K; return band in goldMINARD'S NUMBERS, estimates vary

    Rigor decisions

    • The 1812 map georeferenced poorly (1.18% error), so the route follows the printed symbols in {x, y}: the general axis, not the exact march.
    • Later paintings are labelled as illustrations; three were discarded for showing bodies in the foreground.
    • 422K / 100K / 10K are attributed to Minard; other sources give higher entry figures.
    • Typhus is disputed (Vilnius study), so the narration only says "disease".
    • The tsar's refusal to negotiate was left out: not confirmed in the sources read.

    Sources (excerpt)

    • Fondation Napoléon, Passage of the Berezina: napoleon.org
    • Montana State University, Typhus in Russia
    • Wikipedia: French invasion of Russia; Battle of Berezina; Battle of Maloyaroslavets

    Next steps

    Reply "approved" to start Phase 2: narration, word-synced drawings and captions, music, sound effects, still checks, loudness at −14 LUFS and the final MP4. Sources go in a pinned comment, not on screen.

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

    About this skill

    History and educational content often struggle with retention on social media or rely on low-quality AI-generated imagery. This skill automates the production of high-retention documentary shorts using Remotion, focusing on real historical maps and verified public domain archives rather than generic 3D models.

    What it does

    • Source Verification searches official archives and public domain databases for historically accurate maps and photos.
    • Map Georeferencing aligns historical scans to real-world coordinates for precise tactical and territorial drawing.
    • Automated Animation draws battle lines, routes, and labels directly onto historical materials, synced to narration word-for-word.
    • Retention Engineering optimizes the first 3 seconds with specific hook structures and "Frame 0" covers designed for Instagram and TikTok algorithms.
    • Narrative Logic structures scripts around causal chains and tension rather than simple chronological lists.

    How it works

    1. Research & Asset Selection uses fetch_media.py to find verified assets and georef.py to prepare historical maps.
    2. Phase 1 Delivery provides a complete script (roteiro.md), chronological scene plan, and source list for approval.
    3. Phase 2 Implementation generates high-quality TTS, copies the Remotion engine template, and binds animations to specific spoken words.
    4. Validation & Export runs TypeScript checks, renders stills for visual audit, and exports the final MP4 with optimized loudness.

    Frameworks & tools

    Built on the Remotion video-as-code framework. It utilizes Python scripts for asset fetching and geocoding, Edge-TTS for audio generation, and FFmpeg for final media processing and metadata extraction.

    Why this beats prompting it yourself

    Generic prompts often result in hallucinated historical facts or static, boring slide shows. This skill enforces strict retention metrics, uses real-world map coordinates for drawing, and records the source and licence of every image and checks facts against two sources.

    Use cases

    • Creating viral history Reels explaining specific battles or revolutions.
    • Producing educational Shorts about engineering feats like the Panama Canal.
    • Developing short-form biographical documentaries for social media growth.

    Known limitations

    Does not support AI image generation for historical scenes to maintain rigor. Final delivery is a local MP4 file; it does not auto-post to social media platforms.

    Uses real archive images only, no AI-generated scenes. Narration and on-screen text default to Brazilian Portuguese (it follows the language you ask in). Rendering runs locally and needs Node, Python and ffmpeg; delivery is a local MP4, nothing is posted. Remotion requires a paid company licence for some businesses. Facts are checked against two sources, but you should still review the script before approving.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Accounts
    GitHubWikimedia
    Software
    Node.js 20+Python 3.12+FFmpegcreate-videoremotiontsc
    Optional
    ElevenLabs API key

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files
    Environment Variables

    Allowed Hosts

    raw.githubusercontent.com
    api.elevenlabs.io{path}
    commons.wikimedia.org
    Listedtoday

    What's inside

    Frequently Asked Questions

    It never generates images. It downloads real historical maps and public-domain images, georeferences each map with control points and draws routes, fronts and regions on top. When a map georeferences poorly, it draws over the printed symbols instead and says so in the script.

    The SKILL.md workflow, the Remotion engine template (map stage with camera, drawings, badges, word-by-word captions, audio), Python scripts for media search and licences, georeferencing, narration, music, sound effects, stills and loudness, references on archives, licences and Reel retention, and a complete example project (images and audio are downloaded or generated, not bundled).

    A coding agent that can run shell commands (e.g. Claude Code), plus Node.js 20+, Python 3.12+ with numpy, scipy, soundfile, pillow and edge-tts, ffmpeg, and the official Remotion agent skills. Tested on macOS (Apple Silicon).

    No, this skill intentionally excludes AI image generation to maintain historical rigor, relying instead on verified public domain archives and official database fetches.

    The skill utilizes 'Retention Engineering' to automate the first 3 seconds with high-impact hooks and 'Frame 0' covers specifically formatted for the TikTok and Instagram Reels algorithms.

    The current version provides a high-quality MP4 file and a full audit log of sources; however, it does not include direct API posting to social platforms to allow for final manual review.

    New in Game Development