pentest-osint-linkedin

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized activities were detected. The skill's behavior is consistent with its stated purpose of assisting in OSINT research.
  • [DATA_EXFILTRATION]: The skill accesses a local cookies.txt file to authenticate its requests to LinkedIn. This is a standard method for session management in scraping tools. The analysis confirmed that these session tokens are only sent to the official LinkedIn domain.
  • [PROMPT_INJECTION]: The skill processes untrusted external data (LinkedIn HTML). Ingestion points: Network fetching in linkedin_orchestrator.py and profile extraction in linkedin_views.py; Boundary markers: Scraped data is presented to the agent context without specific delimiters or isolation warnings; Capability inventory: Includes file system writing (save_text in linkedin_core.py) and network requests to LinkedIn; Sanitization: Content is parsed via BeautifulSoup and cleaned for excessive whitespace.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 10:28 AM
Security Audit — agent-trust-hub — pentest-osint-linkedin