Skills
Navigate
skills/jasonjgardner/blockbench-mcp-plugin/bun-development/Gen Agent Trust Hub

bun-development

Fail

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill documents installation procedures involving piping remote scripts to a shell (e.g., curl -fsSL https://bun.sh/install | bash). These scripts originate from the official domain of the Bun project, a well-known development tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of applications that process external data, creating an attack surface for indirect prompt injection. 1. Ingestion points: Network requests handled by Bun.serve and file reads via Bun.file in SKILL.md. 2. Boundary markers: None provided in the implementation examples. 3. Capability inventory: Powerful capabilities are documented, including Bun.write for file system modification, bun:sqlite for database access, and bun run for script execution across various sections of SKILL.md. 4. Sanitization: The basic implementation examples do not include input validation or sanitization logic.
Recommendations
  • HIGH: Downloads and executes remote code from: https://bun.sh/install - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 4, 2026, 11:33 AM