task-complexity-classifier

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the typesafe-sdk and python-dotenv Python packages to interact with the TypeSafe AI classification service.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided task descriptions which are sent to an external API for classification. This presents an attack surface where malicious input could attempt to influence the classification results.\n
  • Ingestion points: Task descriptions are read from command-line arguments or standard input in classifier.py.\n
  • Boundary markers: The skill does not use specific delimiters to isolate the task description from the classification instructions.\n
  • Capability inventory: The skill initiates network requests to the TypeSafe AI API using the typesafe-sdk library.\n
  • Sanitization: The input text is not sanitized or validated before being transmitted to the external API.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 10:11 PM
Security Audit — agent-trust-hub — task-complexity-classifier