linkedin-navigation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes untrusted external data from LinkedIn which creates a surface for indirect prompt injection.
- Ingestion points: Uses
mcp__claude-in-chrome__read_pageandmcp__claude-in-chrome__take_snapshotto read data from the LinkedIn feed, messaging, and notifications. - Boundary markers: Absent. The instructions do not define delimiters to separate the retrieved page content from the agent's instructions.
- Capability inventory: Possesses browser automation tools including
mcp__claude-in-chrome__navigate,mcp__claude-in-chrome__click, andmcp__claude-in-chrome__form_input. - Sanitization: Absent. There is no evidence of filtering or sanitizing the retrieved content.
- [Safe Best Practices] (SAFE): The skill explicitly enforces a policy of not handling passwords, instructing the user to log in manually instead. This prevents credential harvesting by the skill itself.
Audit Metadata