linkedin-messaging

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the linkedin-scraper-mcp package from PyPI. This is an external dependency necessary for the skill's LinkedIn integration.
  • [COMMAND_EXECUTION]: The installation and configuration process involves executing shell commands, including pip install, uvx, and Windows cmd /c wrappers.
  • [DATA_EXFILTRATION]: The skill accesses and stores sensitive LinkedIn session information, including browser cookies, in the local directory ~/.linkedin-mcp/profile/. While required for the browser-based automation, this directory contains sensitive authentication materials that could allow account access if compromised.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external LinkedIn profiles and messages which could contain malicious instructions, and it interpolates user input into outgoing communications.
  • Ingestion points: Retrieves profile data, job descriptions, and inbox messages from LinkedIn via MCP tools.
  • Boundary markers: The documentation specifies a mandatory dry-run workflow (confirm_send=False) for message sending to facilitate user oversight.
  • Capability inventory: The skill can read private messages, search the LinkedIn network, and send messages or connection requests.
  • Sanitization: The instructions do not specify explicit content sanitization, relying on the underlying MCP server's handling of LinkedIn data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 10:13 AM
Security Audit — agent-trust-hub — linkedin-messaging