cross-examine

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill instructions are purely behavioral and do not contain any detected malicious patterns, obfuscation, or safety bypass attempts.
  • [NO_CODE]: The skill consists exclusively of markdown documentation and a license file. It does not contain any executable scripts, binaries, or platform configurations that could lead to command execution.
  • [PROMPT_INJECTION]: The instruction to "explore the codebase" if a question can be answered there establishes an indirect prompt injection surface, as the agent may ingest untrusted data from the local environment.
  • Ingestion points: Local codebase files (SKILL.md)
  • Boundary markers: Absent; no delimiters or instructions are provided to ignore embedded commands in the code
  • Capability inventory: File system read access (implied by "explore the codebase")
  • Sanitization: Absent
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 02:06 AM
Security Audit — agent-trust-hub — cross-examine