The Agent Skills Directory

Indirect Prompt Injection Surface: The skill is designed to retrieve and process configuration settings and feature flags from an external service. In scenarios where the configuration source is not fully trusted, this data could potentially contain instructions that influence the agent's actions.
Ingestion points: Data is retrieved via get_configuration_setting and list_configuration_settings methods within SKILL.md and references/non-hero-scenarios.md.
Boundary markers: The provided examples do not include explicit delimiting or instructions for the agent to ignore potentially malicious content within configuration values.
Capability inventory: The skill has the ability to read, write, and delete configuration data, as well as manage feature flags and snapshots using the provided SDK methods.
Sanitization: Content retrieved from the service is processed (e.g., using json.loads) without specific validation or filtering for prompt injection patterns.
External Package Installation: The skill requires the installation of the azure-appconfiguration and azure-identity packages.
The installation is performed using the standard pip package manager as shown in SKILL.md.
These packages are official libraries provided by a well-known service provider, which is consistent with the skill's stated purpose.

azure-appconfiguration-py