Security audit

knowledge-graph-memory

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local in-memory knowledge graph library, and the reviewed artifacts do not show hidden, destructive, network, credential, or persistence behavior.

This appears safe to install as a local library. Users should understand it may store caller-provided data in memory during use, and the short-term memory eviction behavior may remove the oldest-created entry rather than a true least-recently-used entry.

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

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (1)

Intent-Code Divergence

Low

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The comment at L642 says the method will evict the 'least recently used' item, which implies selection based on most recent access time. However, the implementation compares only createdAt timestamps and never tracks or uses a last-accessed field for short-term memory entries, so it removes the oldest-created item instead.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.