Guidde?

KINDAYou can build the valuable core, but the established product has meaningful operational complexity. Pick one narrow promise for v1.

BUILD FIRST

One specific customer type and one primary use case
A vertical slice that completes the main job without manual workarounds
Visible limits, review steps, and a fallback when an external dependency fails

LEAVE FOR LATER

Broad third-party integrations
High-scale collaboration, syncing, and permissions
Automation that can cause harm without human review

RESEARCH NOTES

What this tool actually does.

Checked 2026-09-06

Guidde is an AI-powered screen recorder and video documentation platform that converts narrated screen recordings into step-by-step 'guiddes' for how-to tutorials, SOPs, employee training, process documentation, knowledge-base articles, and product demos. Users record via a browser extension or desktop app, pick a browser tab, app window, or full screen, then walk and talk through the process. Optional 'Magic Mic' transcribes narration, removes silent gaps, pairs steps with clicks and on-screen actions, and renders an editable AI voiceover (the original voice is not stored when used). An editor supports step text, captions/subtitles, image steps, blurring sensitive info, voice and transcript editing, and automatic pan-and-zoom onto clicked areas. Capture includes microphone but not system audio. Plans: Free (10-minute recordings, up to 25 guiddes, no Magic Mic), Pro (unlimited guiddes),

SOURCES

RESEARCHED BUILD PROMPT

Give this to your coding agent.

Build a focused replacement for Guidde: an AI-assisted screen-recording tool that turns narrated recordings into step-by-step documentation videos. Core workflow: (1) the user records screen (tab/window/fullscreen) plus mic while walking through a process; (2) an AI layer transcribes narration, removes silences, splits content into steps aligned to clicks/actions, and generates an editable AI voiceover ('magic-mic' behavior); (3) an editor lets them fix step text, trim or swap audio, add captions, blur sensitive regions, and apply pan-zoom emphasis on click areas; (4) the finished 'guide' renders as a video with a step list and auto-generated cover/end cards, ready to share. User-facing requirements: in-browser recorder with countdown, pause/restart/delete, tab/window/fullscreen picker, mic capture, content-type templates (how-to, SOP, training, KB, demo), and a shareable output page. Technical requirements: getDisplayMedia + MediaRecorder for capture, Whisper-class STT for transcription, silence detection for gap removal, a TTS provider for voiceover (e.g., ElevenLabs or OpenAI), ffmpeg.wasm or server-side rendering for final video, React front end, and lightweight persistence (SQLite/Postgres) for projects. Implementation approach: client-side capture with an async server-side AI pipeline and job queue; store a sanitized transcript plus an edit timeline so users can regenerate voiceover without re-recording. Do not copy Guidde's brand, logo, marketing site, or exact UI; use a distinct name and original styling. Out of scope: system-audio capture, mobile apps, SSO/billing/team gates, real-time collaboration, and fine-tuning AI models. Include a README covering setup, API keys, architecture, and known limitations (browser support, no system audio, STT/TTS cost and latency, imperfect step detection).

Visit the official product page