The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface area because it ingests project-specific data and uses it to perform write operations on experiment configurations.
Ingestion points: The skill retrieves external data such as event schemas, existing metrics, and experiment states using tools like experiment-get, experiment-saved-metrics-list, and read-data-schema (referenced in SKILL.md).
Boundary markers: The instructions do not define clear delimiters or "ignore instructions" markers for the agent when processing the results of these data-fetching operations.
Capability inventory: The skill possesses the capability to modify the remote environment via experiment-update and experiment-saved-metrics-create (referenced in SKILL.md and references/metric-configuration.md).
Sanitization: There are no specific instructions for the agent to sanitize or validate the integrity of event names or metric definitions before incorporating them into experiment-update payloads.

configuring-experiment-analytics