The Agent Skills Directory

[SAFE]: The skill focuses entirely on financial modeling logic, Excel formula construction, and professional formatting standards. No malicious instructions or bypass attempts were found.
[COMMAND_EXECUTION]: The skill instructs the agent to execute a local validation script (recalc.py) to recalculate Excel workbooks. This is a documented functional requirement for ensuring the integrity of the generated financial models.
[INDIRECT_PROMPT_INJECTION]: The skill processes external financial data from sources like SEC EDGAR and user-provided Excel templates. While this creates a potential surface for indirect injection, the instructions emphasize rigorous template mapping and manual verification checkpoints with the user, which are effective mitigations for the intended use case.

lbo-model