using-system-architect

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface exists through documentation processing.
  • Ingestion points: The skill is designed to ingest and assess outputs from the system-archaeologist tool, including subsystem catalogs, C4 diagrams, and architecture reports (referenced in SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat external documentation as untrusted or to ignore embedded instructions within those documents.
  • Capability inventory: The analyzed scripts primarily involve generating markdown documentation and do not include subprocess execution, network operations, or file-writing capabilities that could be leveraged by an attacker.
  • Sanitization: No evidence of input validation or sanitization of the architectural data was found.
  • [SAFE]: No hardcoded credentials, API keys, or private secrets were found in the provided files.
  • [SAFE]: No remote code execution patterns, external script downloads, or unauthorized network activity were identified.
  • [SAFE]: No obfuscation, multi-layer encoding, or hidden Unicode character patterns were detected in the prompt text or metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 04:02 AM