SkillForge: Turn screen recordings into reusable OpenClaw skills

One of the biggest friction points with OpenClaw has been creating skills manually - writing SKILL.md files, figuring out the right CLI commands, testing iteratively.

SkillForge (https://skillforge.expert) solves this by letting you record your screen doing a task, then automatically generating the complete skill from the recording.

The flow is:

1. Record yourself doing any task (browser automation, CLI workflow, etc.)

2. SkillForge analyzes the recording with vision AI

3. It generates a complete SKILL.md with all the steps, commands, and logic

4. Install it in OpenClaw and your agent can now do that task autonomously

This is huge for making OpenClaw accessible to non-developers. Instead of writing code, you just show the AI what to do.

Anyone tried it yet? What skills would you want to create from screen recordings?