skills.rest
shetautnetjer
shetautnetjerCOMMUNITY
Explore all skills by this author →
GitHub

Advanced Vision

Expose local MCP-based primitives for screenshots, input control, and window management.

Updated Mar 15, 2026
Search Tags:#screenshots#mcp#governance#openclaw#gui-automation#advanced-vision
One-click install
npx skills add https://github.com/shetautnetjer/advanced-vision --skill advanced-vision
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Advanced Vision
Source: https://github.com/shetautnetjer/advanced-vision/tree/main
Command: npx skills add https://github.com/shetautnetjer/advanced-vision --skill advanced-vision

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a local, governance-ready substrate for computer-use primitives including screenshot capture, mouse/keyboard control, and window management so agents can operate on a host with controlled rights.

Core Features & Use Cases

  • Local MCP-based primitives: screenshots, input actions, and basic verification flows within a trusted boundary.
  • Deterministic workflows: a repeatable capture → analyze → act → verify sequence with optional dry-run safeguards.
  • Governance-ready scaffolding: room for policy envelopes, provenance, and artifact management as a later phase.

Quick Start

Start the MCP server in a local environment and perform a sample screenshot action in dry-run mode.

Frequently Asked Questions about Advanced Vision

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

FAQPage Schema
How do I automate mouse and keyboard control locally for GUI tasks?▼

MCP-based computer-use primitives enable local screenshot capture, mouse and keyboard control, and window management. They provide a deterministic capture to verification sequence within a trusted local host boundary.

How do I start a screenshot capture workflow in dry-run mode?▼

To start a screenshot capture workflow in dry-run mode, launch the local MCP server environment and trigger a sample action. This executes the deterministic capture and verification sequence without applying actual system changes.

Does this MCP server support governance and audit trails for agent actions?▼

Yes, the MCP server supports governance by providing scaffolding for policy envelopes and audit trails. It includes optional safety features like dry-run mode to ensure controlled artifact handling and action provenance.

Can I use local MCP primitives for window management and GUI automation?▼

Yes, you can use local MCP primitives for GUI automation and window management. They expose a capability layer that allows agents to operate on a single host environment with controlled rights and basic verification flows.

What are the limitations of using a local computer-use toolbox for governance?▼

The limitation is that the policy and governance layer is currently scaffolding for a future phase. While deterministic workflows and dry-run safeguards are active, full policy envelopes and artifact management are not yet implemented.

Related Skills

Explore More Software Engineering Skills→
jeffreykthomasjeffreykthomas

add-ollama-tool

Expose on-host Ollama models to NanoClaw agents via an MCP server.

Community
Intermediate
JMTassyJMTassy

feynman/intent_action_audit

Cross-references intention and action artifacts to identify mismatches for HELEN governance audits.

Community
Advanced
frankxaifrankxai
6

Enterprise Orchestration

Coordinate AI agent teams across domains with governance and compliance policies.

Community
Advanced
jameinghjameingh
2

agent-browser

Drive a browser via Chrome DevTools Protocol to navigate, fill forms, click, and capture screenshots.

Community
Advanced