Skip to content
TrustedSkillsbeta
Back to all skills

Openclaw Audit Watchdog

Checked
prompt-security·vlatest·agents·AGPL-3.0

Continuously monitors open-source LLM deployments for prompt injection vulnerabilities and security risks using automated analysis.

Repository

Install on your platform

Showing Claude Code. One command installs this skill into any of the agents below.

  1. 1

    Run this in your project

    terminal
    $ npx skills add prompt-security/clawsec --skill openclaw-audit-watchdog -a claude-code

  2. Uses the open-source skills CLI and needs Node.js. Add -g to install it for your user instead of this project.

What we know about this skill

Publisher

prompt-security

Taken from the source listing. We haven't verified who controls this account.

Install target

Live repository

Not pinned. Installing fetches whatever the repository holds at the time you run the command.

No human has reviewed this code. The checks below are a static scan of the files at one commit — nobody has run this skill or judged whether it works. A skill runs with whatever access you give your agent, so read the source before you install it.

Automated safety checks

scanned

This skill passed all 6 checks. We read its files at one commit and looked for the things that make a skill dangerous to run — not for whether it works.

  • Declares what it does: passed

    SKILL.md parses and declares a name and a description.

    SKILL.md exists, its frontmatter parses, and it declares a name and a description.

  • No undeclared network calls: passed

    Only contacts known or declared hosts: github.com.

    Every host the skill dials — from its scripts, its config endpoints or the commands it tells the agent to run — is either well-known package and API infrastructure or declared in its own frontmatter. Links in documentation are not counted.

    Hosts contacted

    github.comsource host

  • No obfuscated payloads: passed

    No encoded or obfuscated payloads.

    No base64, hex or compressed blob is decoded and then executed, and no encoded shell commands.

  • No credential access: passed

    Reads no keys, tokens, keychains or credential files.

    Neither the code nor the SKILL.md instructions read SSH keys, cloud credentials, keychains, token stores, browser cookie databases or dotfiles that hold secrets.

  • No pipe-to-shell installers: passed

    No pipe-to-shell or download-and-run installers.

    No `curl … | sh` pattern, in the code or in the instructions, that runs code fetched at run time.

  • Pinned to a commit: passed

    Scanned prompt-security/clawsec at c8faf9de88.

    The exact upstream commit SHA that was scanned is recorded, so the result refers to specific bytes.

Commit
c8faf9de88
Manifest
skills/openclaw-audit-watchdog/SKILL.md
Scanned
22 files, 4,322 lines
This is a static scan, not a review. It reads the skill's files; it does not run them, and it cannot tell you whether the skill is any good or whether the code does what its description says. A skill runs with whatever access you give your agent.
What each check looks for →

About this skill

openclaw-audit-watchdog

What it does

This skill functions as an automated security watchdog for AI agents, continuously monitoring their activities to detect and prevent unauthorized actions or policy violations. It ensures that agent behavior remains within defined safety boundaries by actively auditing interactions in real-time.

When to use it

  • High-stakes environments: Deploy when agents have access to sensitive data, financial systems, or critical infrastructure.
  • Compliance requirements: Use when strict adherence to organizational security policies or regulatory standards is mandatory.
  • Unsupervised operation: Enable during periods where human oversight is unavailable or limited.
  • Behavioral drift detection: Activate when testing new agent capabilities that might inadvertently bypass existing safeguards.

Key capabilities

  • Real-time activity monitoring and logging
  • Automated detection of policy violations
  • Prevention of unauthorized command execution
  • Continuous audit trail generation for security reviews

Example prompts

  • "Activate the watchdog to monitor my data analysis agent before it starts processing customer records."
  • "Set up an audit-watchdog session while I test the new code-writing capabilities of my developer bot."
  • "Enable continuous monitoring for this trading agent to ensure no unauthorized transactions occur during market volatility."

Tips & gotchas

Ensure you have clear, written security policies defined before activating the watchdog, as it flags deviations from these rules. The skill requires active installation and configuration within your specific AI agent environment; it does not operate automatically without setup.