bun-file-io

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill correctly documents standard file system and networking APIs for the Bun runtime. All code snippets use generic placeholder file paths and a documentation-standard placeholder domain (example.com), presenting no inherent risk.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents APIs for reading data from external files and URLs into the agent's context, representing a standard attack surface for indirect prompt injection if applied to untrusted sources.
  • Ingestion points: Bun.file().text(), Bun.file().json(), and fetch() calls in SKILL.md.
  • Boundary markers: None (standard API documentation).
  • Capability inventory: File system read/write, directory manipulation, and network requests via fetch.
  • Sanitization: The documentation focuses on API mechanics and does not implement input sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:00 AM
Security Audit — agent-trust-hub — bun-file-io