The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the shell command mkdir -p to create project-specific directories. The directory name is derived from a 'slug' generated from implementation details. The skill provides explicit sanitization rules (restricting the slug to lowercase, alphanumeric characters, and hyphens) to ensure the safety of the shell command.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from a 'Q&A conversation' to refine implementation prompts and generate plans. This creates an indirect injection surface where content from the conversation could influence agent behavior. • Ingestion points: Conversation history and user-provided implementation requirements (SKILL.md, write-plan.md). • Boundary markers: None; the skill does not use specific delimiters or instructions to ignore potential commands embedded in the Q&A context. • Capability inventory: Directory creation (mkdir) and file write operations (blueprint//plan-.md). • Sanitization: The skill contains explicit instructions to sanitize user-derived slugs before they are used in filesystem operations.

blueprint-generate