The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external Domo dataset schemas and application configurations, creating a potential surface for indirect prompt injection.
Ingestion points: Dataset schema retrieval in the Prerequisites section and application/layout configuration retrieval in Steps 2 and 6 of SKILL.md.
Boundary markers: Absent; the skill does not use delimiters or instructions to isolate external content from the agent's core instructions.
Capability inventory: The skill utilizes community-domo-cli to perform sensitive operations including app creation, card creation, and file uploads.
Sanitization: Absent; there is no logic to sanitize or validate metadata (such as column names) before using it in subsequent automation steps.
[COMMAND_EXECUTION]: The skill relies heavily on shell command execution via an external CLI tool to interact with the Domo platform.
Evidence: Extensive use of community-domo-cli throughout SKILL.md for API-driven tasks like app-studio create, cards create, and layout-set.
[DYNAMIC_EXECUTION]: The skill uses Python scripts to dynamically assemble and modify complex JSON layout configurations at runtime.
Evidence: SKILL.md provides an inline Python script for layout building (Step 6), and the skill includes layout-builder.py for more complex assembly tasks.

app-studio-build