Shadow

What is Shadow?

Shadow is a Mac app that captures what you see, hear, and say, then runs prompts or custom "Skills" on that context. It is designed to work in the background for meetings and on-demand actions, turning spoken or screen-based context into outputs such as meeting notes, follow-up emails, transcriptions, code, or rewritten text.

For meetings, Shadow can auto-start and stop with your call, transcribe the conversation, identify speakers, and capture screenshots of meeting windows. For other tasks, it can be triggered with a keyboard shortcut to read selected text, capture the active screen, listen to voice input, and paste the result into the focused app field or show it in Shadow.

Key Features

• Meeting capture without bots: Shadow runs in the background, detects meetings, and records/transcribes them without joining as a visible bot.
• Autopilot meeting mode: It can automatically detect meetings, record discussions, and structure takeaways so the user does not have to start or stop it manually.
• Speaker identification: Shadow labels speakers in real time to make transcripts and meeting notes easier to follow.
• Custom Skills: Users can run editable prompts for tasks such as meeting notes, BANT breakdowns, action items, follow-up emails, customer feedback summaries, or other prompt-defined outputs.
• Action workflow for any screen: A keyboard shortcut can capture the current screen, selected text, and voice input, then run a chosen Skill on that context.
• Output and delivery options: Results can be pasted into the focused field, shown in Shadow, sent via webhook to tools like Zapier, n8n, or Make, or saved as Markdown files.
• Local-first privacy controls: Transcription happens on the Mac, audio stays on the device, and skills only call LLM APIs when triggered by the user.

How to Use Shadow

Install Shadow on a Mac, then let it monitor meetings in the background or trigger it manually with a keyboard shortcut. For meetings, Shadow detects the call, transcribes the discussion, captures relevant screenshots, and runs a meeting Skill when the call ends.

For other work, open Shadow from anywhere, choose or edit a Skill, and provide context by speaking, selecting text, or letting Shadow capture the active screen. The output can then be pasted into the current app, saved, or routed to another tool.

Use Cases

• Meeting notes and summaries: Capture a call, transcribe it, and turn the discussion into structured notes or a concise summary after the meeting ends.
• Follow-up messaging: Convert meeting context into a follow-up email or reply draft based on what was said and what was shown on screen.
• Sales or discovery call review: Use a meeting Skill such as a BANT breakdown or action-item list to organize call notes into a working format.
• Voice typing and rewrite cleanup: Dictate a message or document and have Shadow turn spoken words into polished text with fillers removed and punctuation added.
• Screen-aware writing and code assistance: Trigger an Action Skill on any screen to draft notes, write code, or transform selected text using both what is visible and what you say.

FAQ

What is a Skill? A Skill is a prompt that Shadow runs on captured context. Skills can be edited or written from scratch.

What is the difference between Meeting Skills and Action Skills? Meeting Skills run after a call and are designed for transcripts, summaries, and follow-ups. Action Skills run on demand from any screen using a shortcut, voice input, selected text, or the active window.

Does Shadow join meetings as a bot? No. The site says Shadow operates in the background and does not join as a bot.

Where is my data processed? Transcription happens on the Mac by default, audio stays on the device, and meeting data is stored locally unless a Skill sends data to an external LLM API or integration.

Can Shadow send results to other tools? Yes. The site mentions webhook output to tools such as Zapier, n8n, and Make, and Markdown export for transcripts and Skill results.

Alternatives

• Meeting transcription tools such as Otter.ai, Fireflies, or Fathom: these are mainly focused on call notes and transcripts, while Shadow combines meeting capture with a broader action workflow on screen and in voice input.
• General AI writing assistants: these help draft or rewrite text, but they usually do not capture meetings, screenshots, and screen context in the same workflow.
• Dictation and voice typing apps: these focus on speech-to-text and text cleanup, while Shadow also adds meeting detection, screen capture, and custom Skills.
• Automation platforms: tools like Zapier or n8n handle workflow automation after data is available, whereas Shadow is the capture and prompt layer that can feed those systems.