The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from multiple sources to provide context for the designer agent.
Ingestion points: The skill reads from .design/projects/{project}/references/ (which may contain URLs and markdown), {PROJECT_PATH}/BRIEF.md, {PROJECT_PATH}/research/INDEX.md, and critique files in {PROJECT_PATH}/critique/.
Boundary markers: There are no explicit delimiters or 'ignore embedded instructions' warnings applied to the ingested content before it is passed to the gsp-designer agent.
Capability inventory: The skill possesses significant capabilities including Read, Write, Bash, and the ability to spawn another Agent.
Sanitization: The process does not describe any validation, escaping, or filtering of the content retrieved from the project files or external references.

project-design