The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources, which presents a surface for indirect prompt injection attacks.
Ingestion points: The skill retrieves external content via mcp__researcher__scrape_page, mcp__researcher__search_and_scrape, and mcp__researcher__news_search as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific instructions to treat the fetched content as untrusted, which could lead the agent to inadvertently follow instructions embedded in web pages or documents.
Capability inventory: The skill provides capabilities to search the web, scrape content, and extract text from YouTube transcripts and file formats such as PDF, DOCX, and PPTX.
Sanitization: No sanitization, filtering, or validation of the retrieved external content is mentioned before it is processed by the agent.

sage