linkedin

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the '@linkedapi/linkedin-cli' package from the official NPM registry to enable LinkedIn automation features.
  • [COMMAND_EXECUTION]: The skill uses the 'linkedin' CLI to execute operations such as searching profiles, sending messages, and creating posts.
  • [PROMPT_INJECTION]: The skill exhibits vulnerability to indirect prompt injection through its data processing functions:
  • Ingestion points: The skill retrieves untrusted content from LinkedIn profiles, messages, and posts using commands like 'linkedin person fetch' and 'linkedin message get' (SKILL.md).
  • Boundary markers: There are no instructions or templates provided to wrap retrieved LinkedIn content in delimiters to prevent the agent from obeying instructions embedded in that data.
  • Capability inventory: The agent can perform significant actions such as sending messages, creating posts, and running custom workflows via 'linkedin workflow run' (SKILL.md).
  • Sanitization: The skill does not describe any validation or filtering of content retrieved from LinkedIn before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 01:30 PM
Security Audit — agent-trust-hub — linkedin