The Agent Skills Directory

[SAFE]: The skill contains no executable code, remote script downloads, or obfuscated content. It is a documentation-only skill providing guidelines for better web research results.
[NO_CODE]: No scripts, binaries, or configuration files are provided alongside the SKILL.md file, significantly reducing the attack surface.
[PROMPT_INJECTION]: The methodology encourages reading full content from external websites, which is an inherent surface for indirect prompt injection. While the skill lacks explicit boundary markers or sanitization for this external data, this is part of the intended research capability and is not considered a malicious finding.
Ingestion points: Web content retrieved via web_fetch.
Boundary markers: Absent.
Capability inventory: The skill guides an agent with web search and fetch tools.
Sanitization: Absent.

deep-research