The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided feature requirements to generate code and execute shell commands, creating an attack surface for indirect prompt injection.\n
Ingestion points: User feature requests (e.g., "Implement user authentication") as described in the "When to Use This Skill" section of SKILL.md.\n
Boundary markers: The skill lacks explicit instructions or delimiters to isolate user-provided requirements during the code generation process.\n
Capability inventory: The skill is granted access to potentially dangerous tools including Write, Edit, and Bash.\n
Sanitization: No sanitization or validation of the user's natural language requirements is performed before they are incorporated into generated file content or command arguments.\n- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform project environment setup and dependency installation.\n
Evidence: SKILL.md Step 5 demonstrates executing npm install to set up the project environment.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the download and installation of external libraries via the standard NPM registry.\n
Evidence: SKILL.md Step 5 recommends installing well-known, industry-standard packages including @tanstack/react-query, zustand, zod, and axios.

feature-builder