The Agent Skills Directory

Structured Local Script Execution: The skill utilizes local Python scripts for data validation and processing. For instance, run_plan.py executes a validation script using subprocess.run. This is a controlled execution of internal code intended to ensure data consistency before generating reports.
Excel Model Manipulation: The apply_model_updates.py and model_diff.py scripts perform read and write operations on Excel workbooks using the openpyxl library. These operations are governed by a configuration file (plan.json) and specific registries, ensuring that only explicitly mapped drivers and outputs are modified.
Automated Placeholder Detection: The system includes logic to scan generated outputs for unresolved tokens like {PLACEHOLDER} or MISSING:. This acts as a quality assurance layer to prevent the delivery of incomplete analysis or authoring artifacts to the user.
Input Validation Surface: The skill processes external financial data (transcripts, filings) through normalized CSV files. While this creates a surface for indirect instructions, the skill mitigates this by enforcing strict source tagging and specific data schemas documented in the reference files.

earnings-deep-dive