The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides a potential surface for indirect prompt injection because it is designed to analyze and interpret code implementation details from a local repository.
Ingestion points: The agent is instructed to use Glob, Grep, and Read tools on the src/ directory to understand implementation details before providing architectural advice.
Boundary markers: There are no explicit instructions for the agent to treat file contents as untrusted data or to ignore embedded instructions found within the scanned code.
Capability inventory: The skill's primary capabilities include codebase navigation and file content analysis through standard repository tools.
Sanitization: Content read from the local codebase is processed without filtering or sanitization. This is a common and generally acceptable risk for code-analysis tools where the primary purpose is to evaluate the user's own source code.

angular-architecture