The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh api) to verify user permissions and bun run to execute project-specific scripts for rendering and uploading elements. These are standard operations for a repository management workflow.
[DATA_EXPOSURE]: The skill checks for the presence of AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY to enable direct uploads. It explicitly instructs the agent to verify these keys without printing their values to the console, adhering to security best practices for secret handling.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to review and process user-provided source code (.tsx) and documentation (.mdx) files. This creates a vulnerability surface where malicious instructions could be embedded within the content the agent is tasked to analyze.
Ingestion points: Processes files at packages/docs/elements/<category>/<slug>/index.mdx and <slug>.tsx.
Boundary markers: No specific delimiters or "ignore instructions" markers are defined for the review phase.
Capability inventory: Includes file system modification, command execution via bun, and GitHub API interactions.
Sanitization: There is no explicit sanitization of the contributor-provided content before the agent processes it for review purposes.

submit-element