makepad-basics

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process local documentation files (./references/app-structure.md and ./references/event-handling.md) to generate responses, which presents an attack surface for indirect prompt injection if these files are manipulated.
  • Ingestion points: Reference files listed in SKILL.md.
  • Boundary markers: Absent; the agent is instructed to incorporate content without specific safety delimiters.
  • Capability inventory: Rust code generation and conceptual explanation.
  • Sanitization: Absent; the skill relies on the existing state of the reference files.
  • [SAFE]: All external resources, including the makepad-widgets crate source and general documentation links, point to official GitHub and Crates.io repositories. Instructions for platform setup (e.g., apt-get install, cargo install wasm-pack) are standard for the development environment and are provided as user-facing documentation rather than automated execution scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 11:43 PM
Security Audit — agent-trust-hub — makepad-basics