The Agent Skills Directory

[SAFE]: The skill provides standard development patterns for the Bun ecosystem and does not contain any malicious instructions, obfuscation, or hidden behaviors.
[SAFE]: It emphasizes security best practices, including mandatory use of prepared statements to prevent SQL injection and the argon2id algorithm for secure password hashing.
[COMMAND_EXECUTION]: Provides patterns for executing shell commands using Bun's native $ template and Bun.spawn. These are documented for legitimate development purposes and use Bun's built-in safety features like auto-escaping for variables.
[EXTERNAL_DOWNLOADS]: Documents standard package management (bun install) and references well-known, trusted libraries such as Hono, Zod, and Sharp.
[INDIRECT_PROMPT_INJECTION]: The skill describes handling untrusted data (HTTP requests, file uploads) but mitigates risks by demonstrating input validation with Zod and explicit directory traversal checks (filename.includes('..')).

bun-dev