wacc-computation
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a well-documented tool for corporate finance calculations. All components (markdown, templates, and CSV data) are standard and do not contain executable instructions or malicious patterns.
- [COMMAND_EXECUTION]: The skill utilizes a local Python script (scripts/calculate.py) for financial arithmetic. Technical analysis of the script confirms it is limited to pure functions using standard library components (dataclasses). It does not perform any system calls, subprocess spawning, file system writes, or network requests.
- [DATA_EXPOSURE]: The skill processes financial parameters and a local peer group data file (evals/files/comparables.csv). There is no evidence of sensitive file access, credential harvesting, or network-based exfiltration logic.
- [PROMPT_INJECTION]: No patterns associated with prompt injection, jailbreaking, or instruction override were detected in the instructions or evaluation files. The evaluation file includes a test for 'adversarial' financial inputs (book vs. market value) to ensure the agent follows professional best practices, which is unrelated to security bypass.
Audit Metadata