linkedin-leads

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from external LinkedIn posts and profiles.
  • Ingestion points: Data is retrieved from external sources using scripts such as @reduck/linkedin.com/get_post, @reduck/linkedin.com/get_post_reactors, and various profile enrichment scripts.
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions embedded in the retrieved LinkedIn content.
  • Capability inventory: The agent can perform Google searches and fetch detailed LinkedIn profile and company information. According to the skill description, it does not perform write actions like messaging.
  • Sanitization: The skill does not describe any sanitization or validation of the retrieved text before it is evaluated for lead scoring.
  • [COMMAND_EXECUTION]: The skill requires the user to run a local bridge command (reduck local --cookies) to facilitate authenticated access to LinkedIn data, which establishes a dependency on an external local process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 09:53 AM
Security Audit — agent-trust-hub — linkedin-leads