linkedin-network-capture-replay

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill instructions and referenced checklist explicitly prohibit the use of literal 'Cookie' headers, 'Authorization' headers, and API keys. It directs the agent to use platform-managed session identifiers ('sessionId') rather than raw credentials.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted network data.\n
  • Ingestion points: Network traffic captured via 'browser_network' (SKILL.md).\n
  • Boundary markers: Absent.\n
  • Capability inventory: Replay capability via 'http_request' (SKILL.md).\n
  • Sanitization: Prescribed redaction and filtering rules in 'references/replay-checklist.md'.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 04:49 PM