The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external requirements and project contexts, which are untrusted data sources.
Ingestion points: The skill accepts a [target] argument and instructions explicitly state to "Examine requirements and existing system context" as the first step in the behavioral flow.
Boundary markers: There are no delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded within the design requirements or system context being analyzed.
Capability inventory: The skill utilizes powerful tools including Write (file system access), Bash (command execution), and the Rube MCP toolset (mcp__rube__RUBE_MULTI_EXECUTE_TOOL), which enables interaction with external services such as Slack, Notion, and Jira.
Sanitization: The instructions do not define any sanitization, validation, or filtering of the input data before it is processed by the design tools or shared with external platforms.

sc-design