The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests untrusted data from project files to determine design requirements, creating an indirect prompt injection surface.
Ingestion points: Design documents in docs/plans/ and existing HTML mockups in docs/mockups/.
Boundary markers: The instructions do not define delimiters or warnings to ignore embedded instructions in these files.
Capability inventory: The skill writes to the file system (docs/mockups/), performs git commit, and invokes the crucible:quality-gate tool.
Sanitization: There is no content validation or sanitization mentioned for the ingested data.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell-level commands and platform tools as part of its workflow.
Evidence: It requires the agent to commit files to git and invoke the crucible:quality-gate command after generating the mockup.

mockup-builder