The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the oodle CLI tool using the oodle-ai/oodle/oodle Homebrew tap. This resource is provided by the skill's author (oodle-ai) and is consistent with the skill's primary purpose.
[COMMAND_EXECUTION]: The skill executes various oodle CLI commands for configuration and management of alerting resources. It also uses jq to process JSON output from these commands. These are legitimate operations within the scope of managing infrastructure components.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with external configuration data retrieved via the oodle CLI, which could potentially contain malicious instructions if the Oodle environment itself is compromised.
Ingestion points: Untrusted data enters the context through oodle notifiers list, oodle notification-policies list, and oodle muting-rules list (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided when the agent processes the JSON results from the CLI.
Capability inventory: The skill has the capability to execute shell commands (SKILL.md) and perform configuration changes via the oodle CLI.
Sanitization: The skill depends on the Oodle server's own validation and error handling for malformed or invalid configuration files.

oodle-alerting