Back to skill

Security audit

Coding

Security checks across malware telemetry and agentic risk

Overview

This is a local coding-preference memory skill with clear limits and no executable or network behavior, though one reference file has confusing wording about updating SKILL.md.

Install only if you want a local coding-style memory in ~/coding/. Review the stored memory periodically, and treat the criteria.md SKILL.md wording as a documentation flaw: preferences should go in ~/coding/memory.md, not into the skill files themselves.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
This skill NEVER:
- Reads project files to infer preferences
- Observes coding patterns without consent
- Makes network requests
- Reads files outside `~/coding/`
- Modifies its own SKILL.md
Confidence
75% confidence
Finding
without consent

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
## Common Traps

- Adding preferences without confirmation → user loses trust
- Inferring from project structure → privacy violation
- Exceeding 100 lines → context bloat
- Vague entries ("good code") → useless, be specific
Confidence
75% confidence
Finding
without confirmation

Self-Modification

High
Category
Rogue Agent
Content
# Criteria for Code Preferences

Reference only — consult when deciding whether to update SKILL.md.

## When to Add
Confidence
85% confidence
Finding
update SKILL

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.