The Agent Skills Directory

[SAFE]: The skill performs routine workspace scanning to identify project frameworks and existing design states. This is a standard initialization procedure for multi-step workflows.
[COMMAND_EXECUTION]: Uses the Bash tool to create necessary directory structures (mkdir) for new style presets. The command parameters are derived from the project's internal style index or user selection.
[DATA_INGESTION]: Ingests data from local files like package.json and .design/config.json to automate setup.
Ingestion points: .design/branding/*/config.json, package.json, .design/system/STACK.md, .design/projects/*/STATE.md.
Boundary markers: Not explicitly defined in the prompts for ignoring embedded instructions in these files.
Capability inventory: Write (file creation), Bash (directory creation), and Skill (invoking sub-skills).
Sanitization: No explicit sanitization or validation of the content of the config files is performed prior to processing.

gsp-start