Sentinel skill

Skill Mickos79/aria-skills/sentinel-skill

Production AI agent skills for OpenClaw — extracted from a real personal AI OS. RAG memory, safety guards, cost tracking, architect agent and more.

Install

npx -y skills add Mickos79/aria-skills --skill sentinel-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

1.3 KB, 310 tokens by cl100k_base, as published. Nobody here has run it

Sentinel Skill — Safe Command Confirmation

Protects your system from destructive operations by requiring explicit confirmation before execution.

What it does

Before running risky commands, the assistant must:

Detect the operation is destructive
Block execution and ask for confirmation
Only proceed after explicit "yes/да/ok/confirm" from the user
Timeout after 60 seconds if no confirmation received

Risky operations that require confirmation

systemctl stop/restart <protected-service>
docker stop/restart/rm <protected-container>
rm -rf, DROP TABLE, DELETE FROM (no WHERE)
git reset --hard, git push --force
Any command touching protected files

Protected services (customize for your setup)

openclaw-gateway, nginx, aria-core, exec-api, n8n

Protected files (never touch)

AGENTS.md, USER.md, SOUL.md, HANDOFF.md, openclaw.json

Implementation pattern

// Before executing any shell command:
if (isSentinelRequired(cmd)) {
  askConfirmation(cmd);  // blocks until user confirms
  return;                // or wait for "да"
}

Rules

Never bypass Sentinel for protected operations
Timeout = 60 seconds (then cancel)
Log all confirmations to audit trail
Trusted commands (git, curl to localhost, node scripts) bypass automatically