ClawHub

Linkedin API

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward LinkdAPI helper for LinkedIn data lookups, but users should treat contact-data queries and generated scripts carefully.

Install only if you trust LinkdAPI and the Python package. Keep the API key in an environment variable, review generated scripts before running them in sensitive workspaces, and use contact-info endpoints only when you have a legitimate, authorized reason to retrieve that data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
85% confidence
Finding
The skill instructs the agent to generate and execute temporary Python scripts via `uv run`, which expands the capability from simple data retrieval to arbitrary local code execution. Even if intended for convenience, this creates a pathway for unsafe script generation, unintended filesystem access, and subprocess misuse if user input is incorporated into generated code or shell commands.

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The workflow normalizes LinkedIn-data retrieval without warning that the requested data may include personal or sensitive profile information. In a skill explicitly designed to fetch people-related data, the lack of privacy guidance increases the risk of collecting or disclosing personal information without appropriate user awareness, authorization, or purpose limitation.

Missing User Warnings

High
Confidence
94% confidence
Finding
The documented `get_contact_info(username)` endpoint explicitly enables retrieval of highly sensitive personal data such as email addresses and phone numbers, yet the skill provides no warning, consent check, or handling restrictions. This materially raises privacy, compliance, and misuse risk because contact details are more sensitive than basic public-profile information and can be abused for targeting, spam, or deanonymization.

VirusTotal

56/56 vendors flagged this skill as clean.

View on VirusTotal