The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to execute several validation scripts in the local environment, specifically python3 scripts/validate_tokens.py, python3 scripts/validate_contrast.py, and python3 scripts/validate_theme_refs.py. It also includes a call to npm run verify to confirm the generated design foundation.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting external data to guide the generation process.
Ingestion points: The agent reads from taste/design-taste.md and taste/aesthetic-systems.md (found in SKILL.md) to infer the brand brief and aesthetic archetype.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the content of these files as data rather than instructions.
Capability inventory: The skill utilizes subprocess execution via Python and npm (found in SKILL.md) which could be exploited if malicious instructions are successfully injected.
Sanitization: No sanitization or validation logic is specified for the input data from the design files before it is processed by the agent.

brandkit