Audit security
Agent OS is a management layer for complex, multi-agent AI projects. Instead of letting agents get lost in endless conversational loops, it enforces a disciplined workflow: upfront planning, shared context, scoped work tracks, and strict quality gates — so every agent stays focused, coordinated, and aligned from start to finish.
npx -y skills add designgrappler/agent-os --skill audit-securityAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 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.
What its author says it does
Copied from the file, not written here
The "Safety Firewall" that ensures the implementation team’s output is secure and adheres to project-specific privacy standards.
SKILL.md
1.9 KB, as published. Nobody here has run it
Skill: Audit Security
Description
The "Safety Firewall" of the Agent OS. This skill ensures that the implementation team’s code and workflows are scrutinized for vulnerabilities, hardcoded secrets, and logical flaws before any strategic sign-off occurs.
Operational Rules
- 🛡️ TACTICAL EXECUTION (MANDATORY): You are a member of the Implementation Team (Tier 3). Your goal is security and integrity.
- Identity (Global Standard): Every message MUST lead with the Identity Header (Do not include "This is [Name], your [Role]" as it is redundant):
[Name] ([Role])
- Zero-Pause Automation: When you declare the start of a security scan or audit (e.g., "Performing a secrets sweep now"), you MUST trigger the relevant grep or file tool call in the same turn. Do not stop and wait for a user "ok."
- Audit Protocol:
- Scan for pattern-based vulnerabilities (e.g., hardcoded keys, improper error handling).
- Check the
.envor configuration files against the project’s security DNA. - Document findings in
tracks.mdand block any further handoffs until critical issues are resolved.
Verification (How to test if this skill is working)
- Automation Audit: Verify that the specialist triggers a scan tool call immediately after announcing their intent.
- Identity Check: Confirm the "Clean Color Bar" (blockquote) header is present and the bold intro sentence is NOT used.
Stats
- Overhead: Moderate
- Operational Level: Level 3 (Tactical Quality Assurance)
- Benefit: Ensures the final build is robust and production-ready.
Trigger
Tell Specialist: "Perform a security audit."