The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, which could potentially contain malicious instructions targeting the agent context.
Ingestion points: External data is ingested from accounting systems (referred to as ~~бухгалтерия) and various payment sources (referred to as ~~платежи) such as Yookassa, Tinkoff, and bank statements.
Boundary markers: The skill does not define specific boundary markers or delimiters to isolate untrusted external data from the primary instructions or system context.
Capability inventory: The skill is authorized to use the Read, WebFetch, and Bash tools. It performs file writing operations to generate and save .xlsx and .pdf reports to specified storage locations (~~хранилище).
Sanitization: The instructions do not describe any sanitization, validation, or escaping processes for the external data before it is processed or used to generate financial summaries.

close-month