h5p-author

Pass

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @missing-elements/h5p-libraries and @zip.js/zip.js packages via a package manager. It also uses npx to run the @missing-elements/h5p-verify tool. These resources are either well-known libraries or official tools provided by the skill's author.
  • [COMMAND_EXECUTION]: The workflow involves running a local Node.js script (scripts/build-quiz.mjs) to package the quiz and using npx to validate and preview the generated .h5p files. These commands are necessary for the tool's core functionality and operate on local files within the project directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to transform user-supplied educational material into H5P quiz data, which introduces a risk of indirect prompt injection.
  • Ingestion points: User-provided source material is processed by the AI to generate the quiz.json source file.
  • Boundary markers: The process uses a structured JSON schema (references/quiz-format.md) as an intermediate step, which helps isolate raw input from the packaging logic.
  • Capability inventory: The tool possesses the ability to create archives on the file system and execute Node.js/npx commands for building and verification.
  • Sanitization: The compilation script scripts/build-quiz.mjs includes an escape function that converts HTML-sensitive characters into safer entities, mitigating potential cross-site scripting (XSS) or injection attacks within the generated content.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 2, 2026, 09:07 AM