The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes telemetry data from Coralogix APIs, which is an untrusted ingestion point. 1. Ingestion points: Observability data retrieved via cx logs, cx spans, cx metrics, and cx search-fields (SKILL.md). 2. Boundary markers: Absent; the skill does not define specific delimiters or warnings for treating fetched telemetry data as untrusted. 3. Capability inventory: Subprocess execution of the cx CLI tool for querying and field discovery (SKILL.md), and reading local markdown files from the references/ directory. 4. Sanitization: Absent; the skill does not specify filtering or validation logic for the content of the telemetry data.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a vendor-specific CLI tool (cx) for metrics discovery, field search, and log/span querying (SKILL.md, references/logs-querying.md). These commands are defined as read-only operations and are central to the skill's functionality.

cx-telemetry-querying