The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill processes external Excel and CSV files. While it primarily performs metadata analysis and provides routing logic, the extracted content is intended for LLM consumption. Maliciously crafted Excel files could attempt to influence the LLM's behavior during the 'HTML Semantic Mode' extraction phase.
Ingestion points: scripts/complexity_analyzer.py reads user-provided Excel files.
Boundary markers: None found.
Capability inventory: File reading and metadata analysis.
Sanitization: None provided in the analysis script.
[External Downloads] (SAFE): Relies on standard, trusted Python packages pandas and openpyxl for data processing.
[Command Execution] (SAFE): Includes a helper script scripts/complexity_analyzer.py which is used for metadata scanning. The script uses standard file access and does not spawn dangerous subprocesses or execute arbitrary commands.

excel-parser