The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources (Linear tickets) which provides a surface for indirect prompt injection if the content contains adversarial instructions.
Ingestion points: The skill instructions specify searching existing records via /api/v1/search?q=<terms> and reading human answers posted to Linear tickets.
Capability inventory: The skill utilizes the linearis tool for reading and writing to the customer's Linear instance.
Boundary markers: There are no explicit instructions for the agent to treat the ingested ticket content as untrusted data or to use delimiters to separate data from instructions.
Sanitization: The skill does not define any specific validation or filtering logic for the content retrieved from Linear before it is processed by the agent.

ask