Screenshot Capture

Process screenshots Enzo shares with comments. Save to reference library, extract content, categorize, set reminders, and log patterns. Use when Enzo sends an image with context like "save this", shares a screenshot of content (LinkedIn posts, tweets, articles), or sends ideas/frameworks to remember.

This skill looks internally consistent, but check these practical items before installing: - Confirm where the 'notes' directory lives and ensure you want screenshots and extracted text written there (who can read/backup/encrypt those files). - Ask the author how 'extract key content' is implemented (OCR/local tool or external API). If external OCR is used, content could leave your environment. - Clarify how reminders are created (which calendar or task system) and whether they will trigger notifications externally; the skill defaults to a 1-week reminder unless told otherwise. - Ensure the agent has appropriate filesystem permissions and that you’re comfortable with automatic saving of potentially sensitive screenshots (test with non-sensitive images first). - If you don’t want automatic autonomous behavior, consider restricting the skill’s ability to run without explicit user approval or ask for an explicit opt-in step before saving or setting reminders. If these questions are answered to your satisfaction (especially about OCR/reminder destinations and storage/access controls), the skill is proportionate to its stated purpose.

Type: OpenClaw Skill Name: screenshot-capture Version: 1.0.0 The OpenClaw skill bundle instructs the AI agent to perform local file system operations, specifically copying inbound images to `notes/screenshots/` and writing extracted content and metadata to markdown files within a `notes/` directory (e.g., `notes/frameworks.md`, `notes/ideas.md`, `notes/patterns.md`). These actions are directly aligned with the stated purpose of processing, saving, and organizing information from screenshots. There is no evidence of data exfiltration, malicious execution, persistence, or prompt injection with a harmful objective beyond the scope of the skill's stated purpose, as all file operations are contained within designated local paths.