seam-ripper

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional, providing a logical framework for architectural analysis. No malicious behaviors or code execution patterns were identified in the markdown instructions or the associated packaging file.\n- [PROMPT_INJECTION]: No evidence of prompt injection or safety guideline bypass was detected. The instructions are task-oriented and remain within the boundaries of standard codebase interrogation.\n- [DATA_EXFILTRATION]: The skill does not include network calls, credential access, or any patterns suggesting data exfiltration. All analysis is performed within the local context provided by the user.\n- [COMMAND_EXECUTION]: There is no request for system command execution or script running within the skill definition.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 04:49 AM