autoskill

Name: autoskill
Availability: InStock
Author: shushuzn

Discover repeated research workflows from screen activity and draft new skill definitions.

4|Updated Mar 2, 2026

Search Tags:#skill generation #workflow analysis #research automation #clustering #local llm #screenpipe #composition recipes

Ready to run

npx skills add https://github.com/shushuzn/Rairos --skill autoskill-shushuzn

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: autoskill
Source: https://github.com/shushuzn/Rairos/tree/main/skills/autoskill
Command: npx skills add https://github.com/shushuzn/Rairos --skill autoskill-shushuzn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires httpx, pyyaml, sentence-transformers, screenpipe, and includes scripts (resource) and references (resource) components.

What problem does it solve?

You spend time repeating the same research workflow, but your skill library doesn’t reflect how you actually work, so you can’t automate those steps as reusable AI skills.

Core Features & Use Cases

Screen-based workflow mining: Observes your recent on-screen actions via the local screenpipe daemon and converts them into a timeline of apps, windows, and captured text.
Local clustering of repeated patterns: Groups events into sessions and clusters recurring workflow “signatures” without needing any external dataset.
Skill gap discovery and drafting: Compares each cluster against existing scientific-agent skills and drafts new skills or composition recipes for workflows not yet covered.

Quick Start

Ask the AI: "Analyze my last 4 hours of work and propose new skills I should add, using my existing screen activity patterns."

Frequently Asked Questions about autoskill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I discover repeated research workflows from observed screen activity?▼

To discover repeated research workflows from observed screen activity, this skill uses the local screenpipe daemon to build a timeline of apps and windows, then clusters recurring workflow signatures locally to find patterns.

Do I need a local screenpipe API to analyze screen activity patterns?▼

Yes, you need a local screenpipe HTTP API running on loopback at http://localhost:3030 to analyze screen activity patterns, and the skill refuses to run if the screenpipe daemon is unreachable.

How does local LLM clustering match workflows against an existing skill library?▼

Local LLM clustering matches workflows against an existing skill library by using local embedding similarity for top-k candidate selection, comparing clustered session signatures to synthesize reuse decisions or draft novel skills.

Can I draft new AI skill definitions for uncovered workflow patterns automatically?▼

Yes, you can draft new AI skill definitions for uncovered workflow patterns automatically, as the skill synthesizes novel or composed SKILL.md drafts for workflows not yet covered by existing skills.

Does the workflow analysis tool redact sensitive strings before making LLM calls?▼

Yes, the workflow analysis tool redacts sensitive strings before making any LLM calls, ensuring private data from your screenpipe timeline is sanitized during skill gap discovery and drafting.

What happens if the screenpipe daemon is unreachable when generating skills?▼

If the screenpipe daemon is unreachable when generating skills, the process refuses to run entirely, as it requires continuous access to the local screenpipe HTTP API on loopback to capture screen activity.