← Back to Skills Marketplace
135
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install openweruh
Description
Proactive screen context processor. Triggered when screen.context payload arrives via webhook. Analyze the visual content, apply the configured persona mode,...
Usage Guidance
This skill appears coherent for its stated purpose but asks you to trust how screenshots/context will be handled in practice. Before installing, ask the provider/developer: (1) Where is weruh.yaml expected to live and what will the agent read from it? (2) Where are screen captures and derived observations stored (local path, DB, remote service)? Are they encrypted and how long retained? (3) Which channel(s) will notifications be sent to, and what credentials or webhooks does that integration require? (4) Is webhook traffic validated and authenticated to prevent spoofed events? (5) Request a privacy/security summary or source code (or a homepage) so you can inspect storage/telemetry behavior. If you cannot confirm safe storage and access controls for captured screen content (which can contain sensitive data), avoid enabling the skill or restrict it to test accounts with non-sensitive data.
Capability Analysis
Type: OpenClaw Skill
Name: openweruh
Version: 1.0.0
The skill bundle contains instructions for an AI agent to act as a screen context processor for productivity and research assistance. It lacks executable code and relies entirely on Markdown instructions (SKILL.md) to guide the agent's behavior based on screen capture data. The instructions are consistent with the stated purpose of providing proactive notifications (e.g., focus nudges or research suggestions), and there is no evidence of data exfiltration, malicious execution, or harmful prompt injection.
Capability Assessment
Purpose & Capability
Name/description (proactive screen context processor) align with the instructions: analyze incoming screen.context payloads and decide whether to notify the user. The skill declares no binaries, env vars, or installs — which is proportional for an instruction-only processor.
Instruction Scope
Instructions stay within the declared purpose (analyze images/text and optionally notify). Two ambiguous items: (1) the instructions reference reading a configuration file 'weruh.yaml' if available — the path and permissions are unspecified (this implies potential local file access), and (2) 'record the context for potential future daily summaries' and mention of a scheduled cron job imply storage of observations but do not specify where/how long or whether images/text are persisted or transmitted. These are implementation details to clarify before installation.
Install Mechanism
No install spec and no code files — instruction-only skill. Low install risk (nothing is downloaded or written by the skill itself).
Credentials
The skill requests no environment variables, credentials, or config paths. That matches the described capability. Note that operational use (webhook delivery of images and posting notifications) will require external wiring (webhook source and destination) not declared here; those integration credentials are not part of this skill and should be provisioned separately and carefully.
Persistence & Privilege
always:false and no explicit persistence requested in the skill. However, the docs mention summarizing observations at day-end via a scheduled job and referencing a local config ('weruh.yaml'), which suggests the system that implements the skill may store observations/config. Confirm where observations are stored, retention, and access controls before enabling.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openweruh - After installation, invoke the skill by name or use
/openweruh - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
OpenWeruh 1.0.0 – Initial release
- Proactive screen context processor skill triggered by screen.context webhook events.
- Analyzes visual or text context from the user's active display.
- Applies persona modes: skeptic, researcher, focus, guardian, silent.
- Determines whether to send user notifications based on configurable logic for each mode.
- Interventions begin with [👁️ OpenWeruh] and are concise, relevant, and mode-specific.
- Default behavior assumes guardian or silent mode if user configuration is missing.
Metadata
Frequently Asked Questions
What is Openweruh?
Proactive screen context processor. Triggered when screen.context payload arrives via webhook. Analyze the visual content, apply the configured persona mode,... It is an AI Agent Skill for Claude Code / OpenClaw, with 135 downloads so far.
How do I install Openweruh?
Run "/install openweruh" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openweruh free?
Yes, Openweruh is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Openweruh support?
Openweruh is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openweruh?
It is built and maintained by fikriaf (@fikriaf); the current version is v1.0.0.
More Skills