skills/dyrits/skills/classify/Gen Agent Trust Hub

classify

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses curl to send data to https://classifier.dev for text classification. This is the primary purpose of the skill and targets a dedicated service domain. The instructions specifically warn against sending secrets or private information to this third-party API, recommending local classification or redaction instead.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use curl for API interactions. It follows best practices by recommending --data-binary @- and passing data via standard input to prevent command injection through shell argument substitution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external text for classification (ingestion point). It mitigates risks by recommending the use of structured JSON for API requests and instructs the agent to report only the final classification verdicts rather than raw data or JSON, which reduces the surface area for downstream injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 10:49 PM