Back to skill

Security audit

Paid Bundle V1.1

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed production-agent operations guide, with persistent memory features that should be enabled deliberately.

Install only if you want agent operations guidance that may enable long-term memory and multi-agent workflows. Review any additional phase files before use, and only add the memory environment variable to shell or OpenClaw config if you are comfortable with agent memory persisting across sessions.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

Quick Install

bash
# Create skills directory
mkdir -p ~/.openclaw/workspace/skills/production-agent-ops

# Copy all 7 SKILL.md files (downloaded from this package)