Skills
skills/famaoai-creator/gemini-skills/text-classifier/Gen Agent Trust Hub

text-classifier

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it ingests and processes content from external files provided via the --input argument.
  • Ingestion points: The argv.input path is used by classifyText in src/lib.ts to identify file content for classification.
  • Boundary markers: The skill does not implement explicit delimiters or safety warnings for the LLM regarding the content of the file being processed.
  • Capability inventory: The skill uses @agent/core's classifier.classifyFile to analyze the file and KnowledgeProvider.getJson to load rule sets.
  • Sanitization: The ruleSet parameter is strictly validated against a whitelist of 'doc-type', 'domain', and 'intent'. No content-level sanitization is performed on the input file itself before classification.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 02:29 AM