ClawHub
Back to skill

Security audit

Claw Memory Manager

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed OpenClaw memory configuration helper with operational and privacy implications, but no hidden or purpose-mismatched behavior was found.

Before installing, use --dry-run to inspect planned changes, avoid the aggressive Active Memory preset unless you are comfortable with broader memory recall, and use --no-restart in managed or production environments where gateway restarts require coordination.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding
The skill advertises and documents capabilities to read environment configuration, write configuration files, and trigger a shell-based gateway restart, but there is no declared permissions section warning users or the host about those sensitive actions. This is dangerous because users and orchestration systems may invoke the skill expecting a low-risk documentation-style action, while it can modify persistent agent behavior and operational state.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill description emphasizes convenience and safety features, but it does not prominently warn that using the skill changes OpenClaw configuration files and may automatically restart the gateway. This can lead to unsafe invocation in production or managed environments where a restart or config mutation has availability and change-control consequences.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documentation states that the skill automatically writes configuration and triggers a gateway restart, but it does not present this behavior with an explicit operational warning or user-consent emphasis. In a configuration-management skill, silent or default service restarts can cause unexpected disruption, apply unintended settings broadly, and increase the blast radius of misuse or operator error.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
Enabling active-memory changes the agent to proactively retrieve and inject prior conversational/contextual data before each turn, which can expose sensitive information across interactions if users are not clearly warned at the moment of enablement. In a memory-management skill, this behavior is expected functionality, but the privacy impact is still real because the action persists a configuration change and can broaden data exposure without explicit informed consent.

Static analysis

No suspicious patterns detected.