The Agent Skills Directory

[SAFE]: The skill facilitates character creation within a local development workflow, strictly managing data through predefined JSON schemas and project paths. It does not perform any sensitive system operations or network requests.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection due to its processing of project artifacts.
Ingestion points: Data enters the process via load_artifact calls targeting analysis and rewrite_seed files in SKILL.md.
Boundary markers: No specific delimiters or safety instructions are used to separate ingested artifact content from the skill's operational instructions.
Capability inventory: The skill utilizes MCP tools to write JSON data to the projects/ directory and log decision audits.
Sanitization: The skill does not implement explicit validation or escaping of the ingested artifact content.

character-design