The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to fetch and process content from untrusted external websites (portfolio links provided by users).
Ingestion points: The agent is instructed to fetch and read home pages, About pages, and case studies from user-provided URLs (SKILL.md, Section 2).
Boundary markers: The instructions lack explicit delimiters or guidance to ignore potentially malicious instructions embedded in the fetched portfolio content.
Capability inventory: The skill uses network access to fetch URLs and may utilize a browser to access content that fails a simple fetch (SKILL.md, Section 2.1).
Sanitization: There is no logic provided to sanitize or filter the content retrieved from external sites before the agent analyzes it.
[EXTERNAL_DOWNLOADS]: The README.md file provides instructions for manual installation that involve downloading the skill's markdown file directly from a remote GitHub repository using curl.
[COMMAND_EXECUTION]: The README.md includes a shell command snippet (mkdir -p ... && curl ...) intended for the user to execute in their terminal to set up the skill environment.

portfolio-review