ClawHub
Back to skill

Security audit

public-api-finder

Security checks across malware telemetry and agentic risk

Overview

The skill is mainly a coherent public API search helper, but it also recommends uploading research artifacts to an external durable storage service without enough consent or retention guidance.

Install only if you are comfortable with the API search tool making network requests and caching public API data locally. Do not allow Vaultline uploads unless you explicitly want the research saved externally; avoid storing private project details, prompts, credentials, internal URLs, or proprietary notes there unless you understand the service’s retention and access controls.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill invokes external CLIs and scripts that can access the network, read environment variables, and potentially read/write files, yet it declares no permissions or warnings about those capabilities. This creates a transparency and policy-enforcement gap: a caller may treat the skill as low-risk while it can perform actions with broader side effects, including consuming secrets from the environment or persisting data indirectly.

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The Vaultline section introduces durable storage and exfiltration-adjacent behavior unrelated to simply finding public APIs. Because it uploads research content using an API key from the environment and encourages persistence across agents, it expands the skill's data handling surface and can cause sensitive prompts, project context, or research artifacts to be stored externally without necessity for the core task.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The example explicitly recommends uploading research artifacts to Vaultline for durable reuse but does not warn users that the data will persist outside transient chat history. In agent settings, this can lead to unintentional retention of sensitive project details, queries, URLs, or internal notes, and the context makes it more dangerous because persistence is framed as a convenience rather than an opt-in risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.