The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill is well-structured for its intended purpose of financial data organization.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to process external financial data from connectors and user-uploaded CSV/Excel files. However, the risk is negligible as the skill lacks exploitable capabilities like arbitrary command execution, network exfiltration, or file system modifications.
Ingestion points: Data ingested via accounting and payment connectors (e.g., ~~бухгалтерия) and user-provided CSV fallbacks as described in SKILL.md and reference/connector-queries.md.
Boundary markers: The instructions do not define specific delimiters for separating data content from instructions, though they advise the agent on specific fields to extract.
Capability inventory: Analysis of all referenced files confirms no usage of subprocess calls, dynamic code execution (eval/exec), or outbound network operations.
Sanitization: The skill focuses on extracting and formatting financial data into Markdown reports; no specific sanitization logic is implemented for the raw input data.

tax-season-organizer