drama-workflow

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The script scripts/drama_workflow.py performs dynamic path manipulation to import dependencies from the vendor's local file structure (../../text-splitter/scripts and ../../text-truncator/scripts). This behavior constitutes dynamic loading from computed paths, which is functional for a multi-skill workflow but involves modifying the runtime path.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted story content that could contain embedded instructions intended to override agent logic.
  • Ingestion points: The input_text parameter in the execute_workflow method of scripts/drama_workflow.py.
  • Boundary markers: Absent; the skill does not use delimiters to wrap external data or provide 'ignore instructions' warnings.
  • Capability inventory: The script performs text splitting, truncation, and coordinates the integration of analysis results into a structured report.
  • Sanitization: Absent; no validation, escaping, or filtering is performed on the user-provided story text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:49 PM
Security Audit — agent-trust-hub — drama-workflow