mdbook

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides an installation script (install.sh) and instructions in the README.md to execute it using the curl | bash pattern. This script facilitates the setup of the skill by cloning the repository from the author's own GitHub account, which represents standard vendor functionality for installation.
  • [EXTERNAL_DOWNLOADS]: The scripts/generate-references.sh script downloads the official mdBook guide from the rust-lang/mdBook repository to populate the local references/ directory. The SKILL.md also provides URLs for fetching the latest documentation directly from this well-known and trusted source.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests documentation data from external sources, creating an attack surface for indirect prompt injection. However, the risk is negligible as the data is sourced from the official, highly reputable mdBook project repository. Evidence chain: 1. Ingestion points: references/ directory and remote markdown files from rust-lang/mdBook. 2. Boundary markers: None present. 3. Capability inventory: File reading and local reference lookups. 4. Sanitization: Standard markdown processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 05:36 PM