More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    dictate

    1

    Point, talk, submit. Review a video or your website by talking while you point, and Claude gets notes tied to the exact timecode or page element, with screenshots, then starts the edits.

    Free

    Security scanned
    dictate

    dictate

    Example session with this skill installed

    /dictate the onboarding pages

    • Read your context and instructions
    • Compiled the dictate
    • Generated the UI component

    Dictate Studio opens. You browse the onboarding pages, point at the signup button and say "make this bigger and green, like the one I showed you". After Send, Claude replies with a numbered plan, for example "1. Signup button in components/SignupForm.tsx: larger and green, styled after reference S1", then makes the edits.

    SignupForm.tsx

    TSX · React component

    Generated

    Example file from a real run - the skill writes it into your workspace.

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

    About this skill

    Show Claude what you want instead of typing it. Type /dictate and Dictate Studio opens. Talk while you point at a video or browse your website, then press Send. Claude gets a report where every note is tied to a timecode or a page element, with screenshots, and starts on the edits: first a numbered plan, then the work.

    • Video review: click the spot on the frame, press T, talk, press T again. Drag to box an area, I and O to mark a range.
    • Web review: browse your site page to page. The Dictate Chrome extension pins the element under your cursor with its selector, text and HTML, and takes screenshots with the pins drawn on them.
    • Show what you like: open any other site in the middle of a review and talk there, or drag a screenshot around the part you mean. Dictate tags it Reference, so Claude builds that look into your code instead of searching your code for it.
    • Arabic and English: Whisper transcribes on your own computer, even when you mix both in one sentence.

    Private by design. Everything stays on your computer: no account, no analytics and no server in between.

    What it needs. Node 20 or newer, and Chrome or Edge for web review. The skill runs npx @d7y/dictate@latest. The first review sets up local speech to text by itself (faster-whisper, up to 3 GB, once) and uses live captions until it's ready.

    Made by D7y. Free and open source (MIT): https://github.com/D7y1/dictate

    Web review needs the Dictate Chrome extension, so it works in Chrome and Edge only. Without Python 3.9+, transcripts fall back to live captions. It reviews videos and web pages, not native desktop or mobile apps.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 day ago

    • Free with an Agensi account

    Requires

    Hugging Face
    Npmjs
    Pypi
    Node.js 20+
    Google Chrome or Microsoft Edge
    @d7y/dictate
    Node.js

    Permissions

    Allowed Hosts

    registry.npmjs.org
    pypi.org
    huggingface.co
    localhost

    File Scopes

    project
    /.dictate/sessions
    ~/.dictate

    Runs the open source @d7y/dictate npm package, which starts a local studio on localhost. Recordings, transcripts and screenshots stay on your computer. The Chrome extension sends data only to that local studio.

    Listed1 day ago

    What's inside

    Frequently Asked Questions

    Point, talk, submit. Review a video or your website by talking while you point, and Claude gets notes tied to the exact timecode or page element, with screenshots, then starts the edits.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use dictate on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.