ClawHub
Back to skill

Security audit

Code Delegate

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed code-delegation guide that uses command execution and remote LLM tooling in ways that match its stated purpose, with some setup and privacy points users should understand.

Before installing, review whether your workspace can be processed by a remote LLM CLI and whether API keys should be provided through your normal secret-management process. Run the documented setup only in a project where creating .delegate-toolkit configs, logs, and reports is acceptable.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill includes concrete shell commands that create directories and write configuration files into the local workspace, but it does not clearly warn the user that running these steps will modify files on disk. In an agent-executed context, this can lead to unintended workspace changes, persistence artifacts, or silent configuration drift, especially because the skill also exposes the exec tool.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documents use of remote LLM connectivity and API key configuration, including exporting API keys as environment variables, but does not clearly disclose that code, prompts, and metadata may be transmitted to third-party services. This creates a privacy and secret-handling risk in enterprise environments where repository contents may be sensitive.

Static analysis

No suspicious patterns detected.