The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and synthesize untrusted data from external websites, which creates a surface for indirect prompt injection attacks where malicious instructions on a webpage could influence the agent's behavior.
Ingestion points: The SKILL.md file instructs the agent to perform 'WebFetch Deep Dives' and 'extract full content from promising results', pulling untrusted data into the agent's context.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat fetched content as data rather than instructions, nor are there warnings to ignore embedded commands.
Capability inventory: The skill focuses on synthesis, fact-verification, and reporting, meaning malicious content could potentially manipulate the final research output or credibility assessments.
Sanitization: No mechanisms for sanitizing, escaping, or filtering external content are defined in the provided instructions.

search-specialist