prompt-qc
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, potentially untrusted data such as style rules and source briefs as part of its prompt audit workflow.
- Ingestion points: Workflow step 1 in SKILL.md (core contract, style rules, dramaturgy rules, continuity rules, reference asset rules, prompt packet schema, source brief, and target adapter skill).
- Boundary markers: None explicitly defined in the instructions.
- Capability inventory: Direct modification of prompt defects and execution of a local validation script (scripts/minimax_h3_prompt_validator.py).
- Sanitization: No explicit sanitization or filtering of input content is documented.
- [DYNAMIC_EXECUTION]: The workflow specifies running a local Python script, scripts/minimax_h3_prompt_validator.py, to validate prompts against model-specific requirements. This is a standard validation pattern for tools in this category.
Audit Metadata