The Agent Skills Directory

[Indirect Prompt Injection Surface]: The skill ingests solution zip files, which are external artifacts that could potentially contain untrusted instructions within XML or JSON components.
Ingestion points: Solution zip files are identified and processed in Phase 2 via glob and unzip operations.
Boundary markers: The skill does not implement explicit boundary markers or delimiters when processing the content of these files, relying on structured XML parsing.
Capability inventory: The skill possesses high-privilege capabilities, including file system access, execution of CLI tools (pac, az), and authenticated Dataverse API operations.
Sanitization: The skill incorporates multiple user-in-the-loop confirmation gates (Phases 1, 2, 3, and 5b) which provide human verification before sensitive operations are executed.
[Environment Security Configuration Modification]: The skill has the ability to modify the blockedattachments setting for the target environment using the pac env update-settings command.
This feature is designed to resolve import failures caused by environment-level file type restrictions.
A mandatory consent gate (Phase 5b.3) ensures that no changes to environment security settings are made without explicit user approval after the blocked extensions are identified.
[Authenticated Command Execution]: The skill utilizes the Bash tool to execute commands through established administrative CLI tools like the Power Platform CLI (PAC) and Azure CLI (AZ).
It captures access tokens and environment configuration to perform solution deployments and environment variable updates.
These actions are performed using the existing security context of the user and follow standard Microsoft ALM workflows.

import-solution