Security audit

OpenClaw Memory Distill

Security checks across malware telemetry and agentic risk

Overview

This skill openly organizes conversation history into local memory files, with privacy caveats around scheduled automatic writes.

Install this only if you want the agent to maintain local long-term memory. Review MEMORY.md and memory/ logs periodically, expand sensitivePatterns for your needs, and avoid enabling cron unless unattended writes of conversation-derived summaries are acceptable.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The skill explicitly supports cron-triggered execution that writes conversation-derived memory files directly, without requiring an interactive confirmation at execution time. Even though this is framed as a convenience feature, autonomous persistence of user-derived content can surprise users and create privacy or consent issues, especially when scheduled jobs run without active user awareness.

Ssd 3

Medium

Confidence: 92% confidence
Finding: The core workflow instructs the agent to scan current and recent conversation history, classify information semantically, and persist it into MEMORY.md and daily logs. This creates a real risk of storing private natural-language content, including sensitive personal, business, or contextual details that do not match simple token/password regexes, thereby extending data exposure beyond the current session.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.