yara-rule-authoring
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external malware samples, IOCs, and threat intelligence to create detection rules, creating a potential surface for indirect prompt injection. * Ingestion points: Workflow involves collecting variants and threat intel strings as described in SKILL.md and reference files. * Capability inventory: Instructions include using yr check, yr scan, yr fmt, and yr dump via shell. * Boundary markers: No specific delimiters or warnings for embedded instructions are provided in the prompts. * Sanitization: No sanitization of input strings is mentioned before rule creation.
- [EXTERNAL_DOWNLOADS]: The documentation references and recommends tools and repositories from well-known security researchers and organizations. Repositories include Neo23x0/signature-base, Neo23x0/yarGen, mandiant/flare-floss, and elastic/protections-artifacts. It also provides installation instructions for yara-x via brew or cargo.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for rule validation and scanning, specifically using the yr utility suite and auxiliary analysis tools like yarGen and FLOSS.
Audit Metadata