The Agent Skills Directory

[SAFE]: The skill performs expected Maven profile management tasks without any detected security risks. It follows a structured workflow that includes human-in-the-loop validation for all configuration changes based on project data.
[PROMPT_INJECTION]: The skill ingests Maven profile IDs from project files, representing an indirect prompt injection surface. This is considered safe as the workflow requires explicit user confirmation before any ID is used in configuration updates.
Ingestion points: Profile IDs are read from module metadata in derived.json via the profiles.py script.
Boundary markers: The workflow uses AskUserQuestion to require explicit user confirmation before any ID is used in a configuration update.
Capability inventory: The skill uses execute-script.py to call manage-config and manage-architecture tools.
Sanitization: The profiles.py script validates profile IDs against defined regex patterns for classification purposes.

manage-maven-profiles