The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The migration script turborepo_migrate.py reads and parses external configuration files from the local filesystem, which creates a potential surface for indirect prompt injection.
Ingestion points: The script reads and parses package.json and pnpm-workspace.yaml files within the analyze_workspace and discover_packages methods in scripts/turborepo_migrate.py.
Boundary markers: None present; the script assumes valid JSON and YAML syntax for standard development files.
Capability inventory: The script can create directories and write new configuration files (turbo.json) or modify existing ones (package.json) based on the data it reads.
Sanitization: No specific sanitization or filtering is performed on the script names or package names extracted from the monorepo packages before they are written to the new configuration.
[DYNAMIC_EXECUTION]: The skill implements logic to dynamically generate application source code, components, and configuration files.
Evidence: In scripts/nextjs_init.py, the create_directory_structure and create_config_files methods generate and write React layouts, page components, and configuration files (layout.tsx, page.tsx, tailwind.config.ts) using string templates and user-supplied parameters.
Evidence: In scripts/turborepo_migrate.py, the generate_turbo_config method dynamically constructs a build pipeline for Turborepo based on the script patterns detected in the monorepo's packages.

web-frameworks