ClawHub
Back to skill

Security audit

Code Handover Assistant

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed codebase handover assistant that reads and summarizes project files into onboarding documents, with no hidden install, persistence, credential use, or destructive behavior found.

Install this if you want an agent to inspect a repository and generate Chinese handover documentation. Be aware it may read many project files, run local listing/search commands, delegate analysis on large repositories, and write HANDOVER.md and QUICKSTART.md. For sensitive repositories, confirm the target path and avoid running it on unrelated private code.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill explicitly directs the agent to use `execute_code` to parse repository `.ipynb` files. Even though the sample code only reads JSON, introducing code-execution tooling into a code-handover skill expands capability from read-only analysis to active execution, which can increase risk through tool misuse, environment exposure, or accidental execution against untrusted repository content.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The trigger phrases include broad, everyday requests such as '帮我看看这个项目' and '读懂这个项目', which are likely to match many normal conversations. Over-broad invocation can activate the skill in contexts where the user did not intend a full repository analysis workflow, increasing the chance of unnecessary tool use, scope creep, and accidental exposure of sensitive code context.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.