repo-inta-opnv-rtk-meta

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious code, obfuscation, or unauthorized commands were detected. The skill consists of high-level architectural instructions for repository analysis.- [SAFE]: The skill does not access sensitive system files or private credentials. References to reverse-engineering workflows and tools are consistent with the skill's stated purpose of software analysis.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and 'absorb' logic from external, user-provided repositories, which is an inherent data ingestion surface.
  • Ingestion points: User-provided repositories and documents (SKILL.md).
  • Boundary markers: Absent; no specific delimiters or instructions to ignore embedded commands are provided.
  • Capability inventory: Use of the /background command for autonomous loops and instructions for conditional binary installation (SKILL.md).
  • Sanitization: No sanitization or validation of the ingested repository content is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 01:58 PM
Security Audit — agent-trust-hub — repo-inta-opnv-rtk-meta