il-invoice-organizer
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: A detailed audit of the skill instructions, metadata, and provided scripts shows no security violations or malicious intent.
- [COMMAND_EXECUTION]: The helper script
scripts/categorize_invoices.pyis used for processing invoice data. It uses only standard Python libraries for logic and file I/O. It does not perform any shell command execution or subprocess creation. - [DATA_EXFILTRATION]: There are no network operations or calls to external domains. The skill does not attempt to access sensitive system paths (e.g., credentials, SSH keys) or transmit any data outside the local environment.
- [PROMPT_INJECTION]: The instructions are focused on functional bookkeeping tasks and do not contain patterns intended to bypass safety guidelines or override agent behavior.
- [REMOTE_CODE_EXECUTION]: All code is bundled locally. There are no external downloads or remote script execution patterns.
- [CREDENTIALS_UNSAFE]: No hardcoded credentials, API keys, or secrets were found. The skill correctly instructs users to manage business data without exposing sensitive information.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied invoice data, creating a potential attack surface. However, the risk is minimal as the processing is performed by a script with restricted capabilities (no network or shell access), and the instructions utilize structured formatting to maintain data boundaries.
Audit Metadata