skills/mukul975/anthropic-cybersecurity-skills/building-ioc-defanging-and-sharing-pipeline/Gen Agent Trust Hub
building-ioc-defanging-and-sharing-pipeline
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: Communicates with well-known security services, including VirusTotal and AbuseIPDB, to enrich indicators with threat intelligence.
- [DATA_EXFILTRATION]: Distributes extracted and processed security indicators to external sharing platforms such as MISP and TAXII servers, which is the primary intended function of the skill.
- [PROMPT_INJECTION]: The skill ingests and processes untrusted text input to extract indicators. While it lacks explicit boundary markers for this data, the use of strict regular expression patterns for extraction significantly limits the surface for indirect prompt injection.
- [EXTERNAL_DOWNLOADS]: Includes configuration options to bypass SSL/TLS certificate verification (via the SKIP_TLS_VERIFY environment variable) when connecting to sharing platforms, which is a common practice for internal lab environments but should be used with caution.
Audit Metadata