skills/clawdbot/skills/linkedin/Gen Agent Trust Hub

linkedin

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill contains markdown documentation and metadata only. No executable scripts, binaries, or automated installation files were detected.\n- [PROMPT_INJECTION]: The skill relies on processing data from LinkedIn profiles and messages, creating a surface for indirect prompt injection.\n
  • Ingestion points: Browser snapshots of LinkedIn messaging, notifications, and profile pages via the browser tool.\n
  • Boundary markers: No explicit markers or instructions to ignore embedded commands are provided for processed external content.\n
  • Capability inventory: The skill uses the browser tool to navigate, click, and type within the user session.\n
  • Sanitization: No specific sanitization methods are listed; however, the skill mandates manual user confirmation for all outgoing messages and connection requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 04:38 PM
Security Audit — agent-trust-hub — linkedin