Security audit

Memory Layered

Security checks for vulnerabilities and agentic risk

Overview

The skill is a prompt-only recipe for maintaining local long-term memory files; its privacy risks are visible and purpose-aligned rather than hidden or destructive.

Install only if you want an agent to maintain durable local memory. Before use, decide which categories of personal or project information may be stored, avoid sensitive data unless explicitly needed, periodically review memory/topics and memory/.learnings, and require confirmation before pruning or archiving important entries.

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

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

Findings (5)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill explicitly promotes persistent storage of user preferences and project state across sessions, but it provides no user-facing notice, consent flow, retention disclosure, or guidance on handling sensitive data. In context, this is dangerous because the memory system is designed to accumulate natural-language personal and project information over time, increasing privacy, compliance, and unintended disclosure risk.
Content: No source excerpt is available for this finding.

Ssd 3

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill instructs the agent to persist and repeatedly maintain user preferences and project state as durable memory, which creates a clear natural-language data retention surface. In this context, the six-layer architecture makes the issue more dangerous because it normalizes long-term accumulation, indexing, and retrieval of potentially sensitive information across sessions and even across multiple agents.
Content: No source excerpt is available for this finding.

Ssd 3

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The example memory index explicitly includes personal identity attributes such as username, age, and city in persistent topic files. This is more dangerous than generic memory guidance because it concretely encourages storing personally identifiable information in durable plaintext-like notes, increasing the likelihood of privacy violations, over-collection, and leakage to other sessions or agents.
Content: No source excerpt is available for this finding.

Ssd 3

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The workflow directs the agent to append new information to topic files after every conversation, encouraging continuous accumulation of user-provided data without clear boundaries. In this skill's context, that repeated writeback pattern materially increases long-term privacy risk, propagation of inaccurate data, and exposure if the memory directory is later read by other agents or users.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The skill documents automated deletion and archival behavior for memory artifacts without warning users that information may be removed, archived, or become hard to recover. This creates data-loss and integrity risks because the agent may automatically forget or move records that users or operators expected to remain available.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.