book-architect

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is comprised solely of markdown instructions and templates for book organization. No executable code, scripts, or binaries were found during the analysis.
  • [SAFE]: There are no hardcoded credentials, sensitive file access patterns, or network operations (such as curl or wget) present in the skill files.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for user-provided documents like Book Concept Documents and Market Research Reports (SKILL.md). It employs structural headers in templates (e.g., assets/templates/master-architecture-template.md) to delineate context. The skill's capability inventory is empty of dangerous tools, significantly limiting the potential impact of instructions embedded in source data. No specific input sanitization instructions are included.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 04:40 PM