The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides an interface for the agent to ingest data from external databases, which serves as a potential surface for indirect prompt injection findings.
Ingestion points: Results from database::query and database::row-change (SKILL.md).
Boundary markers: No specific delimiters or safety instructions are mentioned for handling retrieved data.
Capability inventory: The skill has the capability to perform data modifications and schema changes via database::execute and database::transactionExecute (SKILL.md).
Sanitization: The skill documentation recommends using parameterized syntax (? or $1) to sanitize database inputs.
[DATA_EXPOSURE]: The database::listDatabases function is designed to prevent credential exposure by redacting sensitive information from connection URLs (SKILL.md).
[SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill is a legitimate tool for database management within the iii engine ecosystem.

database