The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from multiple external sources, creating an attack surface for indirect prompt injection.
Ingestion points: The seek_assets.py script fetches metadata such as titles, tags, and descriptions from external APIs (Pexels, Unsplash, Pixabay, Iconify). It also uses Playwright to scrape content from various websites including Mixkit, unDraw, and 100font.com.
Boundary markers: The skill does not employ explicit delimiters or 'ignore embedded instructions' warnings when presenting retrieved results to the agent.
Capability inventory: The skill script seek_assets.py has the ability to perform network requests and write files to the local system via the download command.
Sanitization: Metadata and scraped content are printed to stdout with minimal sanitization, which the agent then processes to present to the user.
[EXTERNAL_DOWNLOADS]: The skill is designed to retrieve media assets and font indices from various remote services. These include well-known providers such as Google Fonts and Pexels. The download command in scripts/seek_assets.py allows the agent to fetch content from a URL and save it to a local path.
[COMMAND_EXECUTION]: The skill's workflow requires the agent to execute a Python CLI tool (scripts/seek_assets.py) to perform its primary functions, including listing available sources, searching for creative assets, and downloading files.

assetseeker