Security audit
Security checks for vulnerabilities and agentic risk
The skill appears to be a placeholder DevOps CLI that overstates its capabilities and writes local command history, but I found no exfiltration, destructive behavior, or hidden remote execution.
Review this skill carefully before installing if you expect real Semaphore, Ansible, Terraform, or deployment automation. It appears low-risk from a malware perspective, but its documentation is misleading and the commands are mostly placeholders that write a local history log.
Presenting a placeholder CLI as a full-featured DevOps automation interface creates a trust mismatch that can mislead operators into using it in privileged environments. In security-sensitive DevOps contexts, undocumented no-op or stub behavior can hide missing controls, cause unsafe assumptions, and mask unexpected local side effects.
Presenting a placeholder CLI as a full-featured DevOps automation interface creates a trust mismatch that can mislead operators into using it in privileged environments. In security-sensitive DevOps contexts, undocumented no-op or stub behavior can hide missing controls, cause unsafe assumptions, and mask unexpected local side effects.
A generic command like 'help' can be triggered accidentally or conflict with common assistant workflows, increasing the chance of unintentional invocation of the skill. While low impact on its own, broad trigger names reduce user intent clarity and can become more dangerous when combined with command execution capabilities.
The instruction to 'Run any command' indicates unconstrained command dispatch, which can expose powerful underlying tooling to arbitrary arguments without documented allowlists, guardrails, or confirmation boundaries. In a DevOps skill context, this is especially risky because commands may target infrastructure, credentials, state files, or local systems with significant operational impact.
## Usage
Run any command: `ansible-ui <command> [args]`
---
💬 Feedback & Feature Requests: https://bytesagain.com/feedback
Powered by BytesAgain | bytesagain.com
The manifest says the skill provides a modern UI and powerful API for tools like Ansible, Terraform/OpenTofu/Terragrunt, and PowerShell. In contrast, this script implements a simple local CLI that echoes placeholder status text for generic commands and appends command history to a local file, with no actual integration with those DevOps systems.
The help text presents commands such as 'Run checks', 'Build project', 'Run tests', 'Deploy guide', and 'Clean artifacts' as operational features. However, the corresponding command handlers do not perform those actions; they only echo descriptive text and log the invocation, which contradicts the user-facing documentation of what the commands do.
No suspicious patterns detected.