The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses several Bash script blocks to perform environment detection and project scanning. It checks for technology stacks (React, Vue, SwiftUI, Compose) by inspecting configuration files like package.json, build.gradle, and Package.swift.
[COMMAND_EXECUTION]: A custom function load_skill is defined to dynamically load sub-skills by reading markdown files from the local filesystem. This is used for modular functionality based on the detected project environment.
[DATA_EXFILTRATION]: The skill performs broad searches and content inspection (e.g., grep -rE) across the project directory to identify UI frameworks and animation libraries. While this brings local file data into the agent's context, it is used for local analysis and no external network exfiltration was found.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the user's project files (like package.json or source code) to guide its brainstorming and design process.

paint