The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local shell commands such as rg (ripgrep) and awk to parse and extract metadata from the local API definition file src/api/swagger/Api.ts. These operations are read-only and limited to the project's source directory.
[PROMPT_INJECTION]: While the skill processes local files as input for code generation, it includes explicit validation gates and verification checklists (e.g., in references/completion-checklist.md) to ensure the generated output is type-safe and adheres to project constraints. It does not exhibit patterns of safety bypass or instruction overriding.
[DATA_EXFILTRATION]: The skill analyzes local source code files (src/api/swagger/Api.ts, src/api/api.ts) to understand API structures. No network requests to external or non-whitelisted domains were identified, and no sensitive credentials or personal data are accessed.
[NO_CODE]: The skill consists entirely of instructional markdown and reference files for the AI agent to follow. It does not include or attempt to install any third-party binary executables or external scripts.

pure-admin-crud-generator