Security audit

Multi Agent V101

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill for setting up multiple OpenClaw agents, and its higher-impact suggestions are visible and aligned with that purpose.

Before installing or following the guide, review any cron jobs, backup automation, shared memory, symlinks, and token configuration. These can affect running services or expose data across agents if configured too broadly, so apply them only to intended workspaces and protect bot/API tokens.

A.I.G

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

SkillSpector

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 (2)

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 57)May include surrounding context.
| Ops | Monitoring | @ops_bot | workspace-ops | | Security | Audits | @sec_bot | workspace-security |

Step 2: Create workspaces

bash
mkdir -p ~/.openclaw/workspace-ops/{skills,memory,scripts,state}

Missing User Warnings

Low

Category: Not specified by scanner
Confidence: 83% confidence
Finding: This markdown file includes a cron-based self-healing script that automatically restarts the OpenClaw gateway and Docker services when they appear down. Because automatic restarts can affect system state and active workloads, the skill description should explicitly warn users that enabling this behavior will cause unattended service restarts.
Content: No source excerpt is available for this finding.

Static analysis