Skip to main content
Screenpipe records screen and audio while it is running, then makes that local history available to AI. Start with one useful output, verify the source data, and expand only after you trust it.
Not sure what to try first? Get your first useful result walks through one recorded task, a copyable prompt, and a check that the answer is useful.

​
For consultants and customer work

Track client time

Reconstruct reviewed work blocks and active-time totals without automatic billing

Log remote-support jobs

Try ten prompts for job logs, billing review, service reports, and scheduled AI work.

Write a weekly client report

Combine work, deliverables, decisions, risks, and next steps into one update

Turn meetings into follow-up

Verify the transcript, then draft decisions and action items

Prepare a CRM update

Extract reviewed deal fields without auto-writing guesses into the CRM

​
For focused individual work

Build a daily project brief

Keep decisions, open loops, and the next step in your existing notes or Obsidian vault.

Review your day

Remember accomplishments, blockers, and open loops

Turn browsing into a brief

Recover sources you viewed and build a checked research summary

Remember relationship context

Prepare a private follow-up queue from meetings and messages

Give an AI local memory

Maintain a small, reviewed context file across agent sessions

​
For operations and engineering

Discover repetitive work

Find one repeated workflow worth improving

Capture an SOP

Turn one clean process run into a tested operating guide

Prepare a detailed workflow handoff

Document inputs, decision rules, evidence, exceptions, and acceptance checks.

Reconstruct an incident

Build a fact, inference, and unknown timeline from a bounded window

Monitor one narrow surface

Watch an approved app, window, or website and report a clear no-data state

​
Find the next answer

Choose the outcome you need. Each guide starts with an answer and a copyable prompt, with examples and deeper detail available when you need them.

​
Make the result repeatable

Once a manual report is useful, follow reliable scheduled reports to verify its saved output and handle missing data or broken connections. Review AI usage and controls to choose its frequency and understand stop and auto-run controls.

​
The safe workflow

1

Set the boundary

Decide which apps, windows, websites, people, and hours belong in the workflow. Add exclusions before recording client, employee, financial, health, or personal information.
2

Verify recording

Complete a short test, then search for something you can see or hear. A blank result is a capture problem or a no-data state—not permission for the AI to guess.
3

Run it manually

Use a current Home shortcut, install a scheduled task from Scheduled tasks → Discover, or describe a custom automation under Scheduled tasks → My tasks → NEW. Run it manually before adding a schedule.
4

Inspect the output

Check time totals, names, decisions, and source moments. Keep uncertain items marked as uncertain.
5

Automate only the stable part

Schedule local drafts after several good manual runs. Require review before sending messages, editing a CRM, billing a client, or changing another system.

​
Rules that keep outputs trustworthy

  • Use /activity-summary for active-time totals; do not infer hours by counting frames, OCR rows, or search results
  • Give broad /search requests a time boundary
  • Distinguish no matching data from the activity did not happen
  • Keep source apps, meetings, URLs, or time ranges when a report may be reviewed later
  • Treat AI project labels, workflow names, and suggested next steps as suggestions
  • Verify current external facts from their original source before publishing a research brief
  • Review any external side effect before it happens
Protected local API endpoints require a bearer token when API authentication is enabled. Retrieve it with npx -y screenpipe@latest auth token or from Settings → Privacy → API security, then follow API recipes.

​
Start small

For personal use, begin with daily work review. For a customer or team pilot, use one person, one workflow, one to two weeks, and one agreed output. This makes privacy, accuracy, and value easier to evaluate before expanding.