finance-tracker
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill implements a comprehensive CLI interface (
finance) for managing financial data. It performs standard file operations (reading and writing JSON/Markdown) within the user's home directory (~/.finance-tracker/). - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input via transaction descriptions and search queries.
- Ingestion points: User-supplied expense descriptions in
finance addand search terms infinance searchprocessed inbin/finance.py. - Boundary markers: The skill does not implement specific prompt-level boundary markers for isolating user-provided text when displayed back to the agent.
- Capability inventory: File system access (read/write) restricted to
~/.finance-tracker/. No network or arbitrary code execution capabilities. - Sanitization: Descriptions are cleaned using regular expressions in
lib/parser.py, and data is safely serialized to JSON inlib/storage.py.
Audit Metadata