mdx-sanitizer

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No malicious intent or patterns detected. The skill's behavior is consistent with its function as a DevOps automation tool.
  • [PROMPT_INJECTION]: Documentation of ingestion surface area.
  • Ingestion points: Content enters the agent context through the reading of local MDX files within the repository.
  • Boundary markers: No specific instruction-ignoring delimiters are defined in the sanitization logic.
  • Capability inventory: The skill utilizes Bash, Write, and Edit tools to modify file content and execute scripts.
  • Sanitization: The tool provides syntactic sanitization for MDX compatibility but does not filter for embedded natural language instructions.
  • [COMMAND_EXECUTION]: Execution of local build scripts.
  • The skill uses the Bash tool to run npm commands such as npm run sanitize:mdx and npm run build, which execute local TypeScript and JavaScript logic defined in the repository's scripts directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:01 AM