crossframe-critical

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill is entirely focused on content structure and critical analysis methodology.
  • [PROMPT_INJECTION]: The skill uses specific invocation triggers and does not contain instructions to bypass safety filters or reveal system prompts. It enforces strict output formats and self-check protocols to maintain task focus.
  • [DATA_EXFILTRATION]: There is no evidence of hardcoded credentials, sensitive file path access (e.g., .ssh, .env), or unauthorized network operations. The mentioned 'source checking' refers to manual or tool-assisted fact-checking rather than malicious data transmission.
  • [REMOTE_CODE_EXECUTION]: The skill does not download external scripts or execute shell commands. It operates entirely through text-based prompting and structural guidelines.
  • [COMMAND_EXECUTION]: No privilege escalation or persistence mechanisms (like sudo, chmod, or crontab modifications) are present in the provided files.
  • [OBFUSCATION]: The files are written in cleartext Markdown and YAML. No Base64, hex encoding, zero-width characters, or hidden acrostics were found during the analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 08:36 AM
Security Audit — agent-trust-hub — crossframe-critical