abn-amro
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a local JavaScript file (abn-amro.mjs) that is executed via Node.js to perform data transformation. It manages file system operations using the built-in fs module to read source files and write CSV output.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external transaction data which acts as an untrusted ingestion point.\n
- Ingestion points: Reads transaction data from user-provided .TAB files via fs.readFileSync.\n
- Boundary markers: The skill does not explicitly provide boundary instructions for the resulting CSV data to downstream LLM consumers.\n
- Capability inventory: File read (readFileSync), file write/append (writeFileSync, appendFileSync), and standard output.\n
- Sanitization: Implements CSV cell sanitization by quoting values and escaping existing quotes to prevent data-driven schema breakage.\n- [SAFE]: No malicious patterns such as obfuscation, persistence, or network-based exfiltration were detected. The script uses only built-in Node.js modules and operates locally on files provided by the user.
Audit Metadata