The Agent Skills Directory

[SAFE] (SAFE): The skill is a pure orchestration layer. It uses structured logic to present options to the user and interpolate variables into predefined command strings. No suspicious obfuscation or remote code execution patterns are present.
[Indirect Prompt Injection] (LOW): The skill processes a plan_path provided as input, which represents an external data surface.
Ingestion points: The plan_path variable is interpolated into several commands including /majestic:build-task, /majestic:run-blueprint, and the task-breakdown tool call.
Boundary markers: None provided in the orchestration instructions.
Capability inventory: The skill can trigger external tools like backlog-manager and internal majestic commands.
Sanitization: None detected in the skill's markdown; safety relies on the downstream tools handling the plan file content securely.

blueprint-execution