Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/screenpipe/screenpipe/llms.txt

Use this file to discover all available pages before exploring further.

Cline is an autonomous AI coding agent for VS Code with 30k+ GitHub stars. it supports MCP servers, so you can connect screenpipe to give Cline context about what you’ve been working on across all your apps.

​
setup

1

Install Cline

install Cline from the VS Code marketplace
2

Open Cline settings

  • open VS Code with Cline installed
  • open Cline panel (sidebar icon)
  • click the gear icon to open settings
3

Add screenpipe MCP server

go to MCP Servers section and add:
{
  "mcpServers": {
    "screenpipe": {
      "command": "npx",
      "args": ["-y", "screenpipe-mcp"]
    }
  }
}
4

Reload MCP servers

click the refresh icon to reload MCP servers
5

Test it

ask Cline: “what was I working on this morning?”
ensure screenpipe is running on localhost:3030 before using Cline with screenpipe.

​
usage

once configured, Cline can search your screen history while coding: recall documentation: 
> I was reading documentation about async/await patterns earlier,
> find it and help me apply those patterns here
find error messages: 
> what error messages have I seen in my terminal today?
reference API responses: 
> find the API response format I was looking at in the browser
> and use it to type this function
meeting context: 
> what did we discuss about authentication in standup?
> help me implement those changes

​
plan mode with screenpipe

Cline’s “Plan” mode works great with screenpipe context:
1

Switch to Plan mode

click “Plan” in Cline’s interface
2

Ask for context

> I was researching authentication patterns earlier,
> find what I was looking at and plan how to implement it here
3

Review the plan

Cline will search your screen history and create a plan based on what you were working on
4

Execute

switch to Act mode to execute the plan
this workflow lets you leverage your entire work context, not just the current editor.

​
available tools

screenpipe provides these tools to Cline via MCP:

​
search-content

search screen OCR, audio transcriptions, and input events:
  • filter by content type (all, ocr, audio, input, accessibility)
  • time range filtering
  • app/window filtering
  • speaker filtering for audio
  • full-text search

​
export-video

create video exports from screen recordings.

​
activity-summary

get app usage and activity overview.

​
list-meetings

list detected meetings with context. see MCP server documentation for detailed parameters.

​
example workflows

context-aware refactoring: 
user: I saw a better pattern for error handling in another codebase today, find it and apply it here

Cline: [searches screenpipe, finds pattern, refactors code]
API integration: 
user: find the API documentation I was reading about the /users endpoint and implement a client for it

Cline: [searches screen history, finds docs, writes client code]
debugging with history: 
user: there was a stack trace in my terminal earlier, find it and help me debug

Cline: [searches screen history, finds error, provides solution]
meeting-driven development: 
user: what features did we decide to build in yesterday's meeting? implement them

Cline: [searches audio transcriptions, extracts requirements, writes code]

​
tips for best results

be specific about what you’re looking for: 
✅ "find the React component I was reading about this morning"
✅ "what was the error message in my terminal 30 minutes ago"
✅ "show me the API docs I had open in Chrome"

❌ "find that thing" (too vague)
mention the app when relevant: 
✅ "find what I was looking at in the browser"
✅ "show me terminal output from earlier"
✅ "what was I working on in VS Code yesterday"
use time references: 
✅ "this morning"
✅ "in the last hour"
✅ "yesterday afternoon"
✅ "during the standup meeting"

​
Cline modes explained

modedescriptionuse with screenpipe
Actautonomous codingask Cline to find and implement based on screen context
Plancreate implementation planresearch phase - gather context from screen history first
AskQ&A modeask about what you’ve been working on
recommendation: use Plan mode with screenpipe to gather context, then switch to Act mode to implement.

​
troubleshooting

Cline not finding screenpipe?
  • verify screenpipe is running: curl http://localhost:3030/health
  • check MCP settings in Cline show screenpipe
  • try refreshing MCP servers (refresh icon)
  • restart VS Code
queries returning no data?
  • ensure screenpipe has screen recording permissions
  • check screenpipe has been running and recording
  • try a broader time range
  • verify the time range has data: curl "http://localhost:3030/search?limit=1"
MCP connection errors?
  • ensure Node.js >= 18.0.0 is installed
  • try npx clear-npx-cache and restart Cline
  • check VS Code’s Output panel for error messages (select “Cline” from dropdown)
Cline timing out?
  • reduce the time range you’re querying
  • be more specific with filters (app name, content type)
  • avoid requesting screenshots unless needed
permission errors on macOS?
  • check System Settings > Privacy & Security > Screen Recording
  • ensure screenpipe app is listed and enabled
  • restart screenpipe after granting permissions
still stuck? ask in our discord — we can help debug.

​
resources