The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes a fetch_url tool to retrieve content from user-provided links. This behavior is intended for ingesting research and articles into the knowledge base and includes an evaluation step to verify content relevance before saving.
[COMMAND_EXECUTION]: The skill interacts with an internal database using SQL queries via the query_database tool. These operations are scoped to providing real-time statistics and status updates about the documents and articles.
[PROMPT_INJECTION]: The skill processes untrusted data from external URLs, which represents a surface for indirect prompt injection. The risk is mitigated by explicit instructions requiring user approval for document modifications and a strict policy against file deletion.

query