The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from the Azure environment via Azure Resource Graph queries, including policy assignment names and resource tags. This content is externally controlled and could potentially be used to influence the agent's behavior if malicious instructions are embedded within resource metadata.
Ingestion points: Data is retrieved from the policyresources and Resources tables using the execute_query tool as described in Step 1 and Step 3 of SKILL.md.
Boundary markers: No specific delimiters or instructions are provided to distinguish between data and potential instructions within the query results.
Capability inventory: The skill utilizes execute_query, generate_query, validate_query, and list_budgets to interact with the Azure environment.
Sanitization: The skill lacks explicit sanitization or filtering logic for the resource metadata before it is interpreted by the agent.

cost-guardrails