skills/itechmeat/llm-code/bun/Gen Agent Trust Hub

bun

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides detailed documentation and usage examples for Bun.$ and Bun.spawn, which are the runtime's native methods for executing shell commands and managing child processes.
  • [DYNAMIC_EXECUTION]: Documentation is provided for the cc and dlopen APIs within the bun:ffi module, which facilitate runtime compilation of C source code and interaction with native shared libraries.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes built-in utilities for parsing and transforming untrusted data, including a Markdown parser (Bun.markdown) and an HTML streaming rewriter (HTMLRewriter). These represent an attack surface for indirect prompt injection if used to process attacker-controlled content without proper sanitization.
  • [EXTERNAL_DOWNLOADS]: The documentation covers the use of the Bun package manager (bun install) and project scaffolding tool (bun create) to fetch dependencies and templates from public registries and version control systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 06:43 AM