analyzing-certificate-transparency-for-phishing

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by processing untrusted data from public Certificate Transparency logs and incorporating it into terminal output and generated markdown reports.
  • Ingestion points: External data enters the context through crt.sh API responses in query_crt_sh (Step 1) and real-time Certstream WebSocket messages in start_monitoring (Step 2).
  • Boundary markers: There are no explicit delimiters or instructions to the agent to disregard potentially malicious commands or markdown within the certificate fields.
  • Capability inventory: The skill has file-write capabilities in Step 4, where it creates ct_report_{domain}.md on the local filesystem.
  • Sanitization: The skill does not sanitize certificate metadata (e.g., common_name, issuer_name) before including it in the report, allowing a malicious certificate owner to potentially inject markdown or instructions intended to influence the agent or the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 12:59 AM
Security Audit — agent-trust-hub — analyzing-certificate-transparency-for-phishing