The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install or run the designlang package from the public npm registry using npm install -g designlang or npx designlang. This involves downloading and executing code from a third-party source not maintained by the skill author.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute the designlang CLI and the cat command to read generated output files into the agent's environment.
[PROMPT_INJECTION]: The skill has a significant surface for indirect prompt injection because it scrapes content from external URLs and reads the extracted design data back into the agent's context without sanitization.
Ingestion points: Website content is processed via npx designlang <url> in SKILL.md.
Boundary markers: The skill lacks delimiters or explicit instructions to ignore embedded commands when reading the output file cat design-extract-output/*-design-language.md.
Capability inventory: The agent has access to Bash, Read, Write, and Glob tools across its scripts, which could be misused if the agent obeys instructions hidden within the scraped web content.
Sanitization: There is no evidence of validation or sanitization of the scraped website content before it is presented to the agent.

extract-design