prioritization-frameworks

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration vectors were identified in the skill instructions or associated files. The provided Python script (prioritization_scorer.py) is used exclusively for arithmetic calculations on project items and relies only on standard libraries.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it ingests untrusted data via JSON input files to generate prioritization reports. However, the risk is minimal because the skill lacks dangerous capabilities (such as network access, arbitrary command execution, or file writing) that could be triggered by processed data.
  • Ingestion points: Untrusted data enters the agent context through the --input file processed by scripts/prioritization_scorer.py.
  • Boundary markers: Absent in the script's output, which prints item names directly into the report.
  • Capability inventory: No subprocess calls, network operations, or file system modifications beyond reading input files are present in the provided tool.
  • Sanitization: The tool does not perform string sanitization, but as a decision-support utility, its impact is limited to text generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 03:18 AM