The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using the Azure CLI (az) to authenticate the user and retrieve access tokens required for environment discovery.
[EXTERNAL_DOWNLOADS]: Fetches environment metadata from the official Microsoft Power Apps API (api.powerapps.com). This is a well-known service and the communication is used to populate configuration settings.
[COMMAND_EXECUTION]: Performs filesystem operations including directory creation (mkdir -p) and writing JSON configuration data to local paths like ~/.copilot/mcp-config.json or project-specific .mcp/ directories.
[PROMPT_INJECTION]: The skill processes data from external APIs and local configuration files which could serve as a surface for indirect prompt injection, although the risk is minimized by the use of trusted sources.
Ingestion points: Environment lists from api.powerapps.com (Step 2a) and existing MCP configurations from the local filesystem (Step 1).
Boundary markers: None identified between external data and agent instructions.
Capability inventory: Shell command execution via az and filesystem write access.
Sanitization: The skill filters JSON properties but does not explicitly sanitize strings against adversarial instructions.

mcp-configure