architecture-audit
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes untrusted data from architecture Markdown files.
- Ingestion points: Processes user-supplied architecture files (e.g.,
architecture.md,design.md). - Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or to ignore embedded instructions within the processed data.
- Capability inventory: The skill has limited capabilities, primarily generating markdown reports (
audit.mdandimprovements.md). It does not perform network operations, subprocess executions, or sensitive file system modifications. - Sanitization: Absent. There is no mention of escaping or validating the content of the provided documents before analysis.
Audit Metadata