skills/mattpocock/skills/prototype/Gen Agent Trust Hub

prototype

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill logic involves reading surrounding project code and user prompts to generate prototype artifacts. Ingestion points: User input and existing codebase content (SKILL.md). Boundary markers: Not defined in the instructions. Capability inventory: File writing (HTML, TSX) and command execution via local task runners (pnpm, python, bun). Sanitization: Not explicitly addressed in the provided instructions.
  • [DYNAMIC_EXECUTION]: The skill generates executable scripts and components (HTML, JS, TSX) to be run within the user's environment (SKILL.md, LOGIC.md, UI.md).
  • [COMMAND_EXECUTION]: The agent is instructed to execute local commands to run the generated prototypes, specifically using pnpm, python, or bun (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:53 PM
Security Audit — agent-trust-hub — prototype