ClawHub
Back to skill

Security audit

Api Doc Generator

Security checks for vulnerabilities and agentic risk

Overview

This API documentation skill asks for broad development-tool access, but the access is disclosed and fits its stated purpose of scanning code, generating docs, running mocks, and managing review workflows.

Install only if you want an agent to read API source code, write generated documentation files, run the api-doc CLI, start local mock services, and use team collaboration tokens. Review target paths, output paths, ports, Git remotes, reviewers, and token storage before allowing execution.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The skill exposes exec capability and repeatedly documents command execution workflows, but its security section only generically recommends a whitelist without defining one. In an agent setting, this gap can lead to user-influenced shell execution patterns, expanding the risk of command injection, unintended system changes, or execution of unsafe binaries.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation guidance is broad enough to trigger on generic API-documentation-related requests, increasing the chance the agent invokes a powerful skill with read/write/exec tools in contexts where the user did not intend it. Over-broad routing is dangerous because it can cause unnecessary repository access, file generation, or command execution under ambiguous prompts.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill instructs repository scanning and output-file generation but does not clearly warn that it will read source trees and write artifacts to disk. In an agent environment, missing disclosure undermines informed consent and can lead to unanticipated access to sensitive code or unapproved modifications of the working directory.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The mock/testing and collaboration sections describe starting services, binding to ports, and enabling reviewer/webhook-style collaboration without clearly warning about network exposure or remote communication. This is dangerous because an agent could start locally exposed services or connect to external systems without the user understanding the security and privacy implications.

Static analysis

No suspicious patterns detected.