The Agent Skills Directory

[SAFE]: The skill performs local mathematical calculations on CSV data and outputs a JSON report. No malicious patterns or security risks were identified.
[EXTERNAL_DOWNLOADS]: No external network requests or downloads were detected in the code or documentation.
[COMMAND_EXECUTION]: The skill uses a local Python script provided with the package which utilizes only the Python standard library.
[DATA_EXFILTRATION]: No network access, hardcoded credentials, or sensitive file access patterns were identified.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted CSV data. However, the risk is minimal as the script performs structured data transformation.
Ingestion points: CSV data read in backlog_prioritization_assistant.py via the --input parameter.
Boundary markers: Absent; the resulting JSON contains raw strings from the input.
Capability inventory: Python open, csv, and json for file processing.
Sanitization: Numeric values are converted to floats; identifiers and titles are not sanitized.

backlog-prioritization-assistant