The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, specifically email threads and meeting transcripts, which could contain malicious instructions designed to influence the agent's output or behavior.
Ingestion points: Data enters the context via spark emails, spark thread, and spark meeting --transcript as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore embedded commands within the fetched data.
Capability inventory: The skill uses commands to read contacts, email bodies, meeting notes, and event schedules.
Sanitization: There are no explicit sanitization or filtering steps defined for the external content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill workflow relies on the execution of multiple spark CLI commands to retrieve data.
Evidence: Execution of spark contacts, spark emails, spark thread, spark meetings, spark meeting, spark events, and spark team in SKILL.md.
Context: These commands are native to the Spark platform developed by the vendor and are used according to their intended functionality for reading user data.

spark-recipe-stakeholder-brief