Skills from clawhub.ai/alexunitario-sketch

This skill assembles token-safe LLM prompts with optional memory retrieval and a memory safety valve; implementation is provided as `scripts/prompt_assemble.py`. No security-relevant behaviors detected.

High-risk skill: reads local context and instructs executing shell commands and cron installation (e.g., `echo "*/5 * * * * /home/athur/.openclaw/workspace/skills/context-restore/scripts/auto_context_monitor.sh >> /var/log/context_monitor.log 2>&1" >> ~/.crontab`, `crontab ~/.crontab`, `python scripts/restore_context.py --install-cron`). It also directs network actions (`git clone https://github.com/openclaw/context-restore.git`) and notification calls.