Back to skill

Security audit

Graph Memory

Security checks for vulnerabilities and agentic risk

Overview

This skill is designed to read local graph memory, but it also auto-loads executable JavaScript from an unreviewed workspace path and installs an unpinned native dependency.

Review this skill carefully before installing. Its main feature is local memory search, so results can reveal prior-session context, names, files, projects, and errors. The larger concern is that the packaged script automatically executes a bridge module from an unreviewed workspace path; install only if you trust and control that local graph-memory code path, and prefer a version that packages or verifies the bridge and pins its npm dependency.

Vulnerability Patterns
  • Tool Hijacking and SpoofingModifies or replaces tools so legitimate-looking calls execute attacker logic
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (2)

T07 · Tool Hijacking and Spoofing

Error
Location
scripts/query.js:27
Finding

Automatic Execution of an Unverified External Workspace Module

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:9
Finding

Unpinned Native npm Dependency Installation

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding

The skill declares access to a persistent local SQLite-backed memory database but does not specify any explicit tool scope, permissions, or allowed-tools boundaries. Even though the skill is presented as read-only, it exposes environment-derived workspace state and persistent session memory, which can broaden data access beyond what users may expect and weakens least-privilege controls.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This skill directly queries persistent session-derived memory containing people, files, projects, and prior context, but it does not clearly warn users that invoking it may reveal historical or sensitive information. That omission increases the risk of inadvertent privacy leakage, especially because the skill is optimized for natural-language exploration of accumulated memory rather than narrowly scoped retrieval.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The code reads from a persistent database under the user's home directory, which may contain user-derived memory data. While the top-level comment describes the file as a knowledge-graph wrapper, there is no explicit user-facing warning, prompt, or privacy disclosure in the code about accessing stored user data.

Content

No source excerpt is available for this finding.