The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses standard instructional language to guide the agent in assisting users with the Bun runtime. No attempts to override safety guidelines or bypass system constraints were identified.
[DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were detected. The instructions prioritize using Bun's built-in Bun.password module for secure hashing, reducing the need for external security packages.
[REMOTE_CODE_EXECUTION]: While the skill mentions bun install for package management, it does not include patterns for downloading and executing untrusted remote scripts or piping URLs to a shell.
[COMMAND_EXECUTION]: The skill guides the agent on using the Bun CLI for development tasks like running servers, tests, and builds, which is consistent with the primary purpose of the tool.
[INDIRECT_PROMPT_INJECTION]: The skill presents an inherent attack surface as it is designed to process user requirements for generating code that handles network and file system operations.
Ingestion points: User-provided project requirements and migration requests in SKILL.md.
Boundary markers: Absent. The skill does not provide explicit markers to separate user input from system instructions during code generation.
Capability inventory: The skill encourages generating code using Bun.serve() (network), Bun.file() (file system), and bun install (package management).
Sanitization: Absent. No instructions are provided for sanitizing or escaping user-provided data before interpolating it into generated application code.

bun