The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides detailed instructions for executing the 'molt' CLI binary. It includes usage examples for database connection, storage backend selection (S3, GCS, Azure, or local paths), and migration control flags.
[CREDENTIALS_UNSAFE]: The documentation references the management of sensitive credentials, such as database connection strings and cloud service provider environment variables (AWS_ACCESS_KEY_ID, GOOGLE_APPLICATION_CREDENTIALS). These references are within the expected context of configuring migration tasks and use placeholders for sensitive values.
[INDIRECT_PROMPT_INJECTION]: The skill defines ingestion points for user-controlled data including POSIX regex table filters in SKILL.md and configuration files (types.json). While explicit boundary markers and sanitization are absent in the documentation, these parameters are intended for the 'molt' binary's operation. The capability inventory includes database access and cloud storage operations which are standard for this utility. This behavior is consistent with the tool's primary purpose and does not represent an elevated security risk.

molt-fetch