The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external knowledge documents, creating a potential surface for indirect prompt injection attacks.
Ingestion points: The skill uses MCP tools get_avatar_knowledge_document and list_avatar_knowledge_documents to retrieve content from external records (SKILL.md).
Boundary markers: There are no explicit instructions or delimiters defined to isolate the ingested knowledge content from the agent's primary instructions.
Capability inventory: The skill includes capabilities to manage account state, including create_avatar, update_avatar, and delete_avatar operations, as well as session management via the SDK.
Sanitization: The instructions do not specify any validation, filtering, or sanitization steps for the data retrieved from knowledge documents before it is processed by the agent.

runway-dev-characters