The Agent Skills Directory

[SAFE]: The skill defines a set of UI mapping rules and architectural patterns for transforming business requirements into a functional CRM prototype. No malicious commands, obfuscation, or unauthorized data access patterns were detected.- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where an AI agent ingests data from an external file to determine the application's structure and behavior.
Ingestion points: The skill explicitly instructs the agent to read and process content from DOMAIN.md or .tasks/domain.md to extract entity fields, terminology, and user roles.
Boundary markers: The instructions do not specify any delimiters or safety warnings to prevent the agent from following instructions that might be embedded within the business domain data.
Capability inventory: The skill is scoped to generating React components, configuring UI layouts, and mapping relationships between data entities. It does not utilize high-risk capabilities such as arbitrary shell execution or external network exfiltration.
Sanitization: There are no explicit instructions to sanitize or validate the content of the domain file before it is used to populate UI labels or configuration objects.

crm-builder