malware-scan-yara

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill invokes the yara command-line utility for its intended purpose of file scanning. All provided examples are standard usage patterns for threat hunting.
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill references the community YARA rules repository on GitHub. This is an industry-standard source and is used only for data (rules), not executable binaries.
  • [PROMPT_INJECTION] (LOW): The skill possesses a surface for indirect prompt injection because it processes untrusted data from the local file system.
  • Ingestion points: Files and directories processed at <target-path>.
  • Boundary markers: None identified in the prompt instructions.
  • Capability inventory: Execution of the yara binary and writing of results to disk using tee.
  • Sanitization: No logic exists to sanitize or escape content from scanned files before reporting results to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:48 PM