The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill's primary function is to direct the agent to download code from an external GitHub repository (https://github.com/yhassy/wireframe-skill) using the npx skills add command.
[REMOTE_CODE_EXECUTION]: The installation process involves executing remote code fetched from a community-managed repository. The suggested command uses the -y flag, which enables automated, non-interactive execution that bypasses user confirmation.
[COMMAND_EXECUTION]: The skill uses shell commands, including grep and filesystem existence checks ([ -e ... ]), to inspect the user's environment and determine whether the tool is installed globally or locally.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it ingests untrusted natural language data to generate structured JSON and interactive HTML outputs.
Ingestion points: Processes natural language UI descriptions provided within user prompts as input for wireframe generation.
Boundary markers: No specific delimiters or safety instructions are defined in the skill logic to isolate user input from the generation process.
Capability inventory: The skill executes shell commands for its setup and requires Python 3 for its core generation tasks.
Sanitization: There is no evidence of input validation or sanitization for the UI descriptions before they are processed into interactive HTML files.

wireframe-skill