The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes metadata from the Power Platform that can be modified by users (makers), creating a surface for indirect instructions to be passed to the agent.
Ingestion points: Data enters the agent's context through tools like list_store_flows, get_store_flow, list_store_makers, and list_store_power_apps. Fields such as displayName, description, tags, and owners are specifically identified as containing user-defined content.
Boundary markers: There are no instructions provided in the skill to wrap this external content in delimiters or to ignore potential instructions embedded within the metadata.
Capability inventory: The skill possesses write capabilities, including the ability to update governance metadata and notification rules via update_store_flow, and to change flow states (starting/stopping flows) via the linked set_live_flow_state tool.
Sanitization: The skill instructions do not specify any sanitization, filtering, or validation logic for the external content before it is interpolated into the agent's prompt context.

power-automate-monitoring