The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and process external data from JSON configuration files (e.g., stack.json, lambda-stack.json) using shell redirection.
Ingestion points: celigo stacks create < stack.json, celigo stacks update <id> < stack.json (SKILL.md)
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands within the JSON files are provided to the agent.
Capability inventory: The skill utilizes the celigo CLI for resource creation (create), updates (update), and sensitive token retrieval (token, rotate-token) (SKILL.md).
Sanitization: No sanitization or schema validation steps are explicitly described for the input data before it is passed to the CLI commands.
[COMMAND_EXECUTION]: The skill makes extensive use of the celigo CLI tool to perform administrative tasks. This includes retrieving sensitive authentication tokens via the celigo stacks token <id> command. The instructions include specific security warnings for the user to handle this output carefully and avoid exposing it in the conversation logs, representing intended vendor functionality for resource management.

managing-stacks