The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to execute several shell commands for environment setup and application lifecycle management. These include using Homebrew (brew) for package management, the Azure CLI (az) for registry authentication, and the custom agentdeploy CLI for managing deployments and retrieving logs. These commands are integral to the primary function of the skill.
[EXTERNAL_DOWNLOADS]: The skill guides the user to install the agentdeploy tool via the vendor's official Homebrew tap (elementx-ai/homebrew-tap). It also references official resources from Microsoft Azure and Homebrew. These downloads originate from trusted or well-known vendor-controlled infrastructure and are documented as standard prerequisites.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface because it retrieves and displays untrusted data from the execution environment. Ingestion points: The agent reads external data using commands such as agentdeploy logs, agentdeploy events, and agentdeploy explain as documented in SKILL.md and references/operations.md. Boundary markers: The instructions do not define clear boundary markers or provide the agent with warnings to disregard potential instructions embedded in the logs or event descriptions. Capability inventory: The agent possesses the ability to execute deployment commands (agentdeploy deploy) and software installation commands (brew install) which could be targeted by a malicious payload in the ingested data. Sanitization: No validation or sanitization mechanisms are specified for the output retrieved from the cluster.

agentdeploy-deploy