Meet InkQuest: The AI-Powered Text Adventure Bringing Dynamic Storytelling to E-Ink Displays

The world of DIY electronics and smart home displays has long held a quiet appreciation for electronic paper, or e-ink technology. Celebrated for its low power consumption, high readability under direct sunlight, and paper-like aesthetic, e-ink is frequently chosen for "set-and-forget" applications. Enthusiasts around the globe routinely transform these gentle, monochrome screens into ambient weather stations, minimalist smart home dashboards, and digital picture frames that blend seamlessly into domestic decor. However, while static displays have dominated the hobbyist space, a new project bridges the gap between passive viewing and dynamic interactivity, breathing fresh life into retro-style gaming.

An electronics enthusiast and developer known on Reddit as Mammoth-Grade-7629 has introduced "InkQuest," an inventive hardware project that transforms a compact microcontroller and an e-paper screen into a living, breathing adventure storybook. By pairing edge-computing hardware with modern generative artificial intelligence, the device crafts a text-based Choose-Your-Own-Adventure (CYOA) game on the fly, ensuring that no two playthroughs are ever quite the same.

The project marries the nostalgic appeal of classic text adventures popularized in the dawn of personal computing with the cutting-edge capabilities of modern large language models. Rather than relying on a pre-written script stored locally in device memory, InkQuest continuously generates its narrative in real-time, adapting directly to the player’s choices and creating an infinitely expandable storytelling medium housed entirely within a portable, low-power enclosure.

The InkQuest Turns Your XIAO ESP32-C3 into an Adventure Storybook

At the physical heart of the InkQuest project is the Seeed Studio XIAO ESP32-C3, a remarkably small yet powerful microcontroller board renowned in the maker community for its compact footprint and integrated wireless connectivity. Because the ESP32-C3 features built-in Wi-Fi capabilities, it serves as the ideal bridge between physical user input and cloud-based intelligence.

When a user boots up the device, the microcontroller establishes a connection to a local wireless network and reaches out to Google’s Gemini artificial intelligence model. The AI is prompted to act as a dungeon master or narrator, generating an initial segment of a fantasy adventure alongside a series of distinct narrative paths for the player to consider. To ensure the hardware can parse the incoming data seamlessly, the Gemini model formats both the story text and the available choices using structured JSON data.

Once the microcontroller receives this structured response, it translates the data and renders the text directly onto the attached e-paper display. The aesthetic of electronic paper proves uniquely suited for this application, offering a high-contrast reading experience that mimics the look of printed pulp fiction or ancient parchment without the harsh backlight glare typical of standard LCD or OLED panels.

It Looks Like a Lot of Fun

Demonstrating the project in motion via a video shared on the popular ESP32 subreddit, the creator showcased how intuitive the physical interaction feels. Navigating the sprawling, AI-generated narrative requires minimal hardware: players use a small, tactile rotary encoder—essentially a physical dial—to scroll through blocks of text and cycle down through the available action prompts. Once the desired course of action is highlighted, pressing down on the dial clicks it into place to select the option.

The mechanics of the gameplay loop are elegantly straightforward. As the creator detailed in the community discussion, every user input is packaged and transmitted right back to the Gemini AI via the microcontroller’s Wi-Fi connection. The model processes the player’s decision within the context of the ongoing narrative history, calculates the consequences of the action, and immediately generates the next segment of the story along with a fresh set of choices. This closed-loop interaction creates a remarkably fluid, conversational gameplay experience that defies the traditional limitations of microcontrollers, which are typically restricted by meager onboard storage capacities.

For hobbyists, developers, and retro gaming enthusiasts eager to explore how modern generative AI can be integrated with physical computing, the creator has made the project fully accessible. Those interested in building their own InkQuest device, examining the technical implementation, or reviewing the source code can visit the official GitHub repository maintained by Jeremy Ngai Sew Poh, where comprehensive documentation and all necessary code files are publicly available.