The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to interact with the system by executing the litium-cloud CLI tool. This includes managing environments, deploying applications, and configuring access controls on the Litium platform.
The commands are used to perform infrastructure-as-code operations and administrative tasks within the Litium Cloud ecosystem.
[EXTERNAL_DOWNLOADS]: The skill references several external resources for its functionality, all of which are managed by the vendor or trusted providers.
It fetches documentation and guides from docs.litium.dev.
It utilizes the official Litium NuGet feed (nuget.litium.com) for tool installation.
It directs users to official portals and monitoring tools at portal.litium.cloud and insights.litium.cloud.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes external data to perform high-privilege operations.
Ingestion points: Reads local YAML manifests (litium-platform.yaml), configuration files (.litium-cloud.config, .npmrc), and ignores files (.litiumcloudignore). It also fetches documentation in Markdown format from the vendor's site.
Boundary markers: The skill expects structured YAML for manifests but does not define explicit delimiters to separate user data from instructions in non-structured inputs like SQL scripts.
Capability inventory: The skill can execute CLI commands, read/write files, and perform network operations via the vendor tool.
Sanitization: There are no explicit instructions for sanitizing external content before it is processed by the agent or the CLI tool.
[DYNAMIC_EXECUTION]: The skill uses dynamic execution patterns to perform administrative tasks.
It supports the execution of SQL scripts via the app action command, where the script content is read from a file.
It uses the apply command to declaratively update infrastructure based on the contents of YAML manifests.

litium-cloud-cli