Hierarchical Memory

• Outcome: Review
• Latest audit: May 28, 2026 at 6:37 AM UTC
• Version:
  1.0.1

Overview

The skill is local-only and broadly matches its memory-management purpose, but its helper script can modify unintended files if given unsafe paths.

Use this only if you are comfortable with a local memory tool that creates and modifies persistent workspace files. Before running scripts/add_branch.py, use simple slug-like names and only pass intended memory/domain Markdown files as parents; ideally patch the script to canonicalize paths and reject traversal or non-memory targets.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The project branch flow trusts the user-supplied --parent path and appends content to any existing file under /root/.openclaw/workspace, not just intended memory domain files. This creates an arbitrary file modification primitive within the workspace, which can corrupt configuration, instructions, or other agent-controlled files and exceeds the skill’s documented purpose of organizing memory files.

VirusTotal

64/64 vendors flagged this skill as clean.