ClawHub
Back to skill

Security audit

sn-update

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed updater for SenseNova skills, but it can replace installed sn-* skills and prune older backups.

Install this only if you want an agent to update installed sn-* skills. Prefer the default OpenSenseNova repository unless you trust a fork, and keep separate backups if you need more than the newest three rollback points.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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 (2)

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill instructs automatic pruning of backups to only the newest three buckets, which permanently deletes older recovery points without requiring explicit user confirmation or warning at execution time. In an update workflow that modifies installed skills, this can cause avoidable data loss and weaken rollback capability after a bad or malicious upstream update.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
with a fork URL.

- **First run**: if you want to actually limit blob download, use partial
  clone with `--filter=blob:none --no-checkout`, then sparse-checkout only
  the selected `skills/<name>` paths before copying them. `--filter=blob:none`
  alone does **not** keep the cache small if the full worktree is checked out;
  that checkout will still download most or all needed blobs. It still
Confidence
88% confidence
Finding
The skill allows the user to override the upstream Git URL with an arbitrary fork and then perform clone/fetch/reset operations against it. In a skill whose purpose is to copy upstream files into active agent skill directories, accepting an untrusted repository source significantly increases supply-chain risk and can install attacker-controlled skills into the agent environment.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.