The Agent Skills Directory

[SAFE]: The skill is purely instructional, providing formatting and design patterns for creating .malloynb notebook files.
[COMMAND_EXECUTION]: The skill mentions platform tools such as execute_query and search_malloy_docs. These are intended for data analysis within the Malloy ecosystem and do not represent unauthorized command execution.
[INDIRECT_PROMPT_INJECTION]: The skill describes a workflow for processing and presenting data that may originate from untrusted sources.
Ingestion points: Malloy queries and markdown text are ingested into the notebook structure (described in SKILL.md).
Boundary markers: The skill explicitly requires the use of >>>markdown and >>>malloy delimiters to separate code from narrative.
Capability inventory: Uses the execute_query tool (mentioned in SKILL.md).
Sanitization: The instructions mandate a 'verify-then-assemble' workflow, requiring the agent to run queries via execute_query and explain results to the user before final report generation.

malloy-analysis-report