Lineage Code Mini

• Outcome: Pass
• Latest audit: May 29, 2026 · 8:29 PM UTC
• Version: 0.1.3

Overview

This skill openly stores local interaction/profile data to personalize future agent responses, with setup and privacy risks users should understand before installing.

Install only if you want the agent to keep local interaction and profile data and use it to shape future replies. Review or delete the data directory periodically, avoid recording sensitive conversation content, and consider installing the npm dependency in a pinned local environment instead of relying on the global setup step.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• 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 (7)

Description-Behavior Mismatch

Medium

• Confidence: 92% confidence
• Finding: The setup script conditionally performs a global npm installation, which expands the skill's scope from local data-directory initialization to modifying the host environment and pulling executable code from an external registry. This is risky because install-time scripts in npm packages may execute during installation, and a global install affects the entire system rather than just this skill.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: This code introduces package-management behavior unrelated to the stated lightweight profiling purpose, giving the skill the ability to alter system-wide runtime state and fetch remote code. In the context of an agent skill, that mismatch is more dangerous because users may expect only local profile storage, not privileged environment modification with supply-chain exposure.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill directs persistent collection of interaction history and behavioral profiling without any user-facing consent, retention notice, or privacy controls. That creates a real privacy/security issue because sensitive behavioral metadata may be stored and reused across conversations without the user's awareness.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill instructs the agent to append generated adaptation content into persistent USER.md/SOUL.md files without warning that this permanently alters long-lived memory/state. Persisting inferred behavioral traits into durable prompt/state files can cause hidden, hard-to-reverse profiling and unintended propagation into future sessions.

Ssd 3

Medium

• Confidence: 94% confidence
• Finding: The skill explicitly tells the host agent to persist per-user interaction history, engagement outcomes, and timing patterns across conversations. That is dangerous because it builds durable behavioral surveillance data that can reveal preferences, activity patterns, and inferred traits, increasing privacy risk and misuse potential.

Ssd 3

Medium

• Confidence: 95% confidence
• Finding: Appending behavioral profile content into persistent USER.md/SOUL.md memory creates durable cross-session state that may silently influence future model behavior. This is risky because inferred traits can become sticky, difficult to audit, and may contaminate unrelated future interactions or be exposed through prompt/context leakage.

Context Leakage

High

• Category: Data Exfiltration

Content

## Automatic Adaptation

This skill provides the commands and storage layout. Your host agent still needs to call these commands or wire recording into its turn loop. Installation alone does not automatically capture conversations.

When this skill is loaded, you should:

• Confidence: 87% confidence
• Finding: capture conversation

VirusTotal

66/66 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.