threat-intel-fusion

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and normalize data from various external and untrusted sources, which establishes a surface for indirect prompt injection attacks.
  • Ingestion points: The workflow involves collecting information from OTX, MISP, abuse.ch, VirusTotal, Shodan, vendor blogs, and internal telemetry as specified in the fusion workflow in SKILL.md.
  • Boundary markers: The instructions lack specific delimiting markers or guidance for the agent to ignore potentially malicious instructions embedded within the ingested threat intelligence.
  • Capability inventory: According to the output contract in SKILL.md, the agent is tasked with generating operational security artifacts, including Sigma, YARA, and Suricata rules, as well as SIEM-ready configurations.
  • Sanitization: There are no explicit requirements or steps defined for sanitizing, escaping, or validating the external content before it is processed or written to detection files.
  • [NO_CODE]: The skill contains high-level operational guidelines and metadata but includes no scripts, binaries, or automated execution components.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:20 PM
Security Audit — agent-trust-hub — threat-intel-fusion