The Agent Skills Directory

[SAFE]: The skill provides legitimate instructions for setting up a development loop with CAST AI, following industry standards for DevOps workflows.
[DATA_EXFILTRATION]: The skill communicates with the official CAST AI API at api.cast.ai to manage cluster policies and retrieve savings data. These network operations target a well-known service infrastructure.
[EXTERNAL_DOWNLOADS]: Downloads and installs the castai-agent using Helm from the official castai-helm repository, which is a trusted source for this vendor's software.
[COMMAND_EXECUTION]: Utilizes terraform, kubectl, helm, and jq for cluster management and infrastructure-as-code tasks, matching the skill's stated purpose.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, creating a potential surface for indirect prompt injection.
Ingestion points: External API responses from api.cast.ai are processed using curl and jq in SKILL.md.
Boundary markers: None present in the instructions.
Capability inventory: The skill has access to Bash for executing terraform, helm, and kubectl commands as specified in allowed-tools.
Sanitization: JSON responses from the API are sanitized and filtered using jq before display or further processing.

castai-local-dev-loop