Project Context Manager

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:23 PM UTC
• Version:
  1.0.0

Overview

This skill creates persistent project notes as advertised, with no evidence of hidden execution, exfiltration, or destructive behavior.

Install this only in workspaces where you want persistent AI-maintained project documentation. Review AI_DOC/PROJECT_CONTEXT.md, AI_DOC/AI_memory/, and AI_DOC/AI_FEEDBACK.md before committing or sharing, and avoid letting secrets, API keys, personal data, or sensitive incident details be written there.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The activation triggers are broad enough to auto-apply this skill during ordinary software-development conversations, which can silently impose its logging and file-management behaviors on users who did not explicitly request them. Because the skill persists context and session traces, overbroad activation increases the chance of unintended data collection and unintended file writes.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill advertises recording cognitive processes and session traces but does not require a user-facing privacy notice or consent flow. This creates a risk that user prompts, project details, secrets, or other sensitive material will be written to disk without the user's informed approval.

Ssd 3

Medium

• Confidence: 97% confidence
• Finding: Requiring a new AI_memory file for each task and capturing items like user intent, context links, discoveries, and self-corrections creates a durable natural-language audit trail of potentially sensitive user and project information. In a development context, these notes can easily include proprietary code details, credentials, architecture notes, incidents, or personal data, expanding the breach surface over time.

Ssd 3

Medium

• Confidence: 95% confidence
• Finding: The append-only History and feedback logs encourage indefinite retention of operational details and collaboration records, with no pruning or expiration policy. That design increases privacy and confidentiality risk because sensitive details may remain on disk long after they are needed and can accumulate across sessions.

VirusTotal

62/62 vendors flagged this skill as clean.