mediar-ai/screenpipe
by Various
YC (S26) | AI that knows what you've seen, said, or heard. Records everything you do, say, hear 24/7, local, private, secure
MCP
mediar-ai/screenpipe
Added 1 June 2026
Overview
Screenpipe is a local recording agent written in Rust that captures screen, audio, and microphone input continuously on your machine. It stores all recordings locally for privacy and enables AI to search and act on what you've seen, said, or heard without sending data to external servers.
Best for
Best for
Developers building local-first AI agents that need rich context from user activity without cloud transmission.
Use cases
- Building AI assistants that understand user context from screen and audio history
- Creating productivity tools that search across everything you've done in a session
- Developing accessibility features that react to on-screen content and spoken input
How to use
Install
npx screenpipe recordTools exposed
Source-availableMulti-monitor
Tested with
Claude Desktop, Claude Code, Cursor, Cline, Continue, VS Code, ChatGPT
Notes
Screenpipe is a local recording agent written in Rust that captures screen, audio, and microphone input continuously on your machine. It stores all recordings locally for privacy and enables AI to search and act on what you’ve seen, said, or heard without sending data to external servers.
19,049 stars on GitHub. Last updated 2026-06-01.
Use cases
- Building AI assistants that understand user context from screen and audio history
- Creating productivity tools that search across everything you’ve done in a session
- Developing accessibility features that react to on-screen content and spoken input
Pros
- Runs entirely local with no cloud dependency, keeping sensitive recordings private
- Captures multimodal input (screen, audio, microphone) in a single unified stream
- Open source with active community (19k+ stars) and Rust performance foundation
Cons
- Requires significant local storage and CPU resources for continuous 24/7 recording
- Privacy and legal compliance responsibility falls entirely on the builder deploying it
- Early-stage tool (YC S26) with evolving API and potential breaking changes
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Runs entirely local with no cloud dependency, keeping sensitive recordings private
- Captures multimodal input (screen, audio, microphone) in a single unified stream
- Open source with active community (19k+ stars) and Rust performance foundation
Cons
- Requires significant local storage and CPU resources for continuous 24/7 recording
- Privacy and legal compliance responsibility falls entirely on the builder deploying it
- Early-stage tool (YC S26) with evolving API and potential breaking changes
Open-source & AI alternatives
Swap-in tools that solve the same job. Weigh the trade-offs before you commit.
ContextPulse/contextpulse
Various
Always-on context for AI agents: screen capture, OCR, voice, MCP server. Local-first, AGPL-3.0.
deusXmachina-dev/memorylane
Various
📇 ☁️ 🏠 🍎 - Desktop app that captures screen activity via event-driven screenshots, stores AI-generated summaries and OCR text locally in SQLite, and exposes your activity histor
grooverLab/fable
Various
High-fidelity transcript memory for Claude Code — index every session, recall byte-identical, search, prune, compose. Local-first, stdlib-only, MCP.
inkbytefo/screenmonitormcp
Various
🐍 🏠 🍎 🪟 🐧 - Real-time screen analysis, context-aware recording, and UI monitoring MCP server. Supports AI vision, event hooks, and multimodal agent workflows.
manovagyanik1/clipy-mcp
Various
Clipy is the first agent-readable screen recorder — free, no watermark, for Mac, Chrome, and the web. One share link: humans watch the video, AI agents read the summary, key moment
ostapondo/Plonk
Various
A native, local macOS toolbox for screenshots, OCR, measuring, keep-awake, pointer tools, shortcuts, windows, workspaces, voice and MCP.
Battam1111/omniseek
Various
Your agent seeks what search can't find. A self-hosted perception MCP server that transcribes speech, reads behind logins, sees images and video frames, crosses languages, and reme
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
whisper
Community
Robust Speech Recognition via Large-Scale Weak Supervision
ollama
Community
Get up and running with Kimi-K2.5, GLM-5, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
llama.cpp
Community
LLM inference in C/C++
ananddtyagi/gif-creator-mcp
Various
MCP that allows you to take your video and turn it into a gif.
clanker-records/crompton-network
Various
Machine-native listening platform for C.W.A.'s Straight Outta Crompton. Your agent can listen. For real.
dimpagk92/cellar
Various
Computer Experience Layer — MCP server for native computer use on macOS. Structured perception, reliable actions, continuous awareness.
dmang-dev/mcp-retroarch
Various
MCP server for RetroArch — exposes memory r/w, save state, screenshot, pause/frameadvance/reset via the Network Control Interface
kindly-software/kdb
Various
Kindly Debugger - Time-Travel Debugging for AI Workflows
lanchuske/local-mcp
Various
LMCP — Give your AI native access to your apps. 189 tools across macOS and Windows. Mail, Calendar, Teams, OneDrive, Outlook, Google Drive and more. 100% local.
Mtehabsim/ScreenPilot
Various
Tool that allows the AI to control your device in the same way you do, enabling automation for everything!
NakaokaRei/swift-mcp-gui
Various
MCP server that can execute commands such as keyboard input and mouse movement on macOS
samvallad33/vestige
Various
Vestige gives AI agents sharp memory: a local-first Rust MCP server that reaches backward through time to find the quiet change, decision, or service that caused today’s failure, n
video-creator/ffmpeg-mcp
Various
Using ffmpeg command line to achieve an mcp server, can be very convenient, through the dialogue to achieve the local video search, tailoring, stitching, playback,clip, overlay, co
huseyinstif/klaket
Various
Turn any video into LLM-ready data - transcripts, speaker labels, scenes, subtitles and on-screen text. Self-hostable, keyless, MCP-ready.
ronak-create/FableCut
Various
Zero-dependency browser video editor that AI agents can drive — JSON timeline, MCP + REST, live-reloading UI
teampolarity/cosmos-mcp
Various
MCP server and local sync package for Cosmos. Connects agents, iMessage, browser history, calendar, Claude Desktop, and shell history to a shared personal graph.
Screenpipe
Various
YC (S26) | AI that knows what you've seen, said, or heard. Records everything you do, say, hear 24/7, local, private, secure
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.