The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The procedure in skills/resize-auditor.md instructs the agent to ingest and analyze external project files, creating a potential indirect prompt injection surface. Ingestion points: The agent reads **/Info.plist and **/*.swift files from the user's project. Boundary markers: The instructions lack explicit delimiters or warnings to the agent to disregard instructions found within the data being analyzed. Capability inventory: The agent is directed to use standard search and read tools (Glob, Grep, Read) and specialized development utilities (devicectl, xcrun). Sanitization: There is no defined logic for sanitizing or filtering the content of the project files before the agent processes them.
[COMMAND_EXECUTION]: The documentation and the auditor logic reference various system and developer tools, including grep, glob, xcrun, and devicectl. These tools are cited within their standard operational context for code analysis and hardware interaction during development.

axiom-uikit