The Agent Skills Directory

[NO_CODE] (SAFE): The skill consists entirely of Markdown instructions (SKILL.md and index.md). It does not include any Python scripts, Node.js modules, shell commands, or configuration files that would interact with the host system or external APIs.
[INDIRECT_PROMPT_INJECTION] (INFO): The skill ingests untrusted data in the form of user research interests and data source descriptions. However, because the skill lacks any 'write' or 'execute' capabilities (e.g., no file creation, subprocess calls, or network requests), this ingestion point does not present a valid attack surface.
[EXTERNAL_DOWNLOADS] (SAFE): The documentation contains references to academic websites (e.g., brown.edu, mostlyharmlesseconometrics.com). These are static hyperlinks intended for human reading and are not used by the skill to download or execute code at runtime.

research-ideation