The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to interact with the observe CLI tool. It provides a structured set of commands for querying and managing observability data, which is the primary and intended function of the skill.
[CREDENTIALS_UNSAFE]: The documentation mentions authentication workflows and the location of local configuration files (~/.observe/config.json). When providing examples of manual configuration, it correctly uses placeholders (e.g., YOUR_API_KEY) to avoid hardcoding secrets.
[DATA_EXFILTRATION]: All network-related commands target observeinc.com subdomains, which are the official infrastructure for the skill's vendor. This activity is consistent with the skill's stated purpose of pulling and pushing telemetry data to the Observe platform.
[PROMPT_INJECTION]: The skill's instructions are focused on guiding the agent's use of the CLI and do not contain any patterns intended to bypass safety filters or override system instructions.

observe-cli