Tldraw Skill

Use when user requests diagrams, flowcharts, architecture charts, or visualizations. Also use proactively when explaining systems with 3+ components, complex...

This skill appears to do what it says (generate tldraw diagrams) but has two behaviours you should consider before installing: 1) Auto-update: it will silently run `git pull` in its install directory (network access and the ability to change the skill code without further consent). If you prefer control, install from a pinned commit, disable the auto-update step (remove or edit it), or run it only in a project-level clone you control. 2) Vision self-check: the skill uses the agent's vision capability to read exported PNGs to auto-fix layout issues — those images may be sent to the model provider for processing. Don’t use the self-check on diagrams that contain sensitive data, or disable the vision step if privacy is a concern. Additional precautions: install the skill in an isolated environment (project-local path), inspect the repository source before enabling auto-update, and prefer explicit manual updates. No credentials are requested by the skill, which is good; still review exported files and where they are written before sharing.

Type: OpenClaw Skill Name: tldraw-pro-skill Version: 1.1.0 The skill is designed to generate and export whiteboard diagrams using the legitimate `@kitschpatrol/tldraw-cli`. It contains standard operational commands for dependency installation (npm), directory management (mkdir), and a documented auto-update feature (git pull) restricted to its own directory. All behaviors are transparently described in SKILL.md and README.md, with no indicators of malicious intent, data exfiltration, or unauthorized system access.