The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface where untrusted content from a user's MongoDB database (specifically schema structures, index definitions, and query results) enters the agent's context through tools like collection-schema, collection-indexes, and aggregate. The skill also has write capabilities such as create-index. To mitigate the risk of indirect prompt injection, the skill defines a strict workflow that requires the agent to explain all technical recommendations in plain language and obtain explicit human approval before execution.
[SAFE]: No evidence of malicious behavior was found. The skill does not contain hardcoded credentials, obfuscated code, persistence mechanisms, or unauthorized network operations. The conditional logic used for MongoDB version checks is a standard technical requirement for feature compatibility and does not constitute a time-delayed attack vector.

mongodb-search-and-ai