Back to skill

Security audit

rpm-ostree Toolkit — Fedora Atomic / Bazzite Manager

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward rpm-ostree command reference for immutable Linux systems, with high-impact system commands that are disclosed and purpose-aligned but should be used carefully.

Before installing or using this skill, review any command that changes deployments, rebases to another image, installs an RPM from a URL, changes kernel arguments, cleans up old deployments, or reboots the system. Verify the image or package source, confirm hardware and branch compatibility, keep a known-good deployment pinned when appropriate, and avoid running high-impact commands unless they match your exact recovery or update goal.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

Rebase changes the operating system image source and can effectively switch distributions/branches, with significant consequences for compatibility, trust roots, package layering, and bootability. In this skill, the commands are presented as straightforward examples without strong warnings to verify remote authenticity, branch suitability, hardware support, backups, and rollback strategy, which makes unsafe rebases more likely in agent-assisted use.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Installing RPMs directly from remote URLs introduces supply-chain risk because the source may be untrusted, tampered with, or unofficial, and the skill does not advise validating provenance, signatures, or HTTPS domain trust before use. In an agent context, users may copy-paste or execute such commands with elevated privileges, turning a documentation shortcut into a realistic path for system compromise.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This section documents rollback, pin/unpin, override reset, kernel argument changes, and cleanup operations that can materially alter boot state or remove recovery points, but it does not consistently warn that these actions require elevated privileges, may reduce rollback safety, or should be confirmed against the user's exact goal first. In an agent skill, omission of guardrails is dangerous because an automated assistant may present or execute these commands as routine remediation, increasing the chance of service disruption or loss of a known-good deployment.

Content

No source excerpt is available for this finding.