Published skills
agent-guardrails
Prevent AI agents from secretly bypassing your rules with mechanical enforcement using git hooks, secret detection, deployment verification, and import registries. This solution, born from real production incidents like server crashes and token leaks, works with Claude Code, Clawdbot, and Cursor for perpetual enforcement.
upgrade-guard
Safe OpenClaw upgrades with snapshot, pre-flight checks, controlled upgrade steps, post-verification, and emergency rollback. Never lose your working system to a bad update again.
config-guard
Prevent OpenClaw config changes from crashing the gateway. It includes auto-backup, schema validation, critical field checks, and auto-rollback, and should be used before any config.apply, config.patch, or openclaw.json edit.
token-guard
Monitor and control OpenClaw token usage and costs. Set daily budgets, track spending, auto-downgrade models when limits hit. Stop burning money while you sleep.
Category alert