The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted financial data from external sources, creating a surface for indirect prompt injection attacks.
Ingestion points: Data is ingested via the /api/v1/aggregate API endpoint and from files (JSON, CSV).
Boundary markers: There are no specified delimiters or instructions to the model to ignore potential commands embedded in the financial data.
Capability inventory: The skill includes capabilities to write files (PDF, CSV, JSON) and perform network operations via API endpoints.
Sanitization: The documentation does not describe any sanitization or validation processes for the ingested data.
[COMMAND_EXECUTION]: The skill utilizes a CLI tool and Python libraries to perform its tasks.
Evidence: Mentions the openclaw CLI tool for data aggregation and the openclaw_financial Python library for report generation and analysis.

financial-reporting