Project Memory Ledger

• Outcome: Review
• Latest audit: May 29, 2026 at 9:26 PM UTC
• Version:
  0.1.3

Overview

The skill is mostly a disclosed project ledger, but Drive mode can write ledger contents to Google Workspace despite local-only wording, so users should review the data flow before installing.

Install only if you are comfortable with a durable project memory ledger. Use backend=local unless you intentionally want Google Workspace writes, check the config paths and Drive folder/doc IDs before running commands, and avoid storing secrets or sensitive internal notes in entries that could be synced to Drive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill clearly instructs execution of python and optional gws-based Drive operations that imply file read/write, shell execution, and network access, yet it declares no permissions. This creates a transparency and policy-enforcement gap: operators may authorize or review the skill as low-risk while it can modify local ledgers, scaffold directories, and interact with external Drive resources.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 85% confidence
• Finding: The skill description emphasizes a local Markdown ledger, but the body also describes broader behaviors: project registration, local scaffolding, PRD proposal generation, and optional Drive folder/document creation and append behavior. This mismatch is dangerous because users and automated reviewers may underestimate the operational reach of the skill, especially the external-write and project-structure side effects.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: In drive mode, project names, purposes, and document contents are transmitted to Google Workspace without a prominent runtime disclosure or confirmation. In a memory-ledger skill, entries may contain sensitive internal decisions or evidence, so silent remote sync can cause unintended data exposure to third-party cloud storage.

VirusTotal

64/64 vendors flagged this skill as clean.