bx-sites-build

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a set of commands to interact with the bxSites CLI tool for site generation (build), local hosting (serve), and environment validation (doctor).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided Markdown content from the project's source directories. While this is an ingestion surface, the risk is mitigated by the tool's architecture.
  • Ingestion points: Markdown files in docs/ and src/ (SKILL.md).
  • Boundary markers: None defined.
  • Capability inventory: Shell command execution (bxSites), local file system writes to site/, and local server hosting.
  • Sanitization: The skill mentions the required bx-esapi and bx-markdown modules, which handle the security encoding and rendering of content.
  • [SAFE]: The skill adheres to security best practices for CLI wrappers and provides specific commands for quality assurance and diagnostics without performing any suspicious network or system operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 03:14 AM
Security Audit — agent-trust-hub — bx-sites-build