The Agent Skills Directory

[PROMPT_INJECTION]: The skill establishes a systematic workflow for ingesting and processing external research papers, creating a potential vector for indirect prompt injection.\n
Ingestion points: The workflow relies on library_search to read paper content and monitor_get_context to scan external feeds or documents for synthesis.\n
Boundary markers: The SOP does not include instructions for using delimiters or explicit system-level isolation when the agent processes untrusted text from external papers.\n
Capability inventory: The skill possesses the capability to write to the local workspace via workspace_save, manage and link tasks via task_create and task_link, and update paper metadata via library_update_paper.\n
Sanitization: The workflow lacks explicit validation or sanitization steps for content extracted from external sources before it is incorporated into syntheses or reports.

Survey SOP