maker-framework
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of Markdown documentation and technical references. No executable scripts or configuration files that trigger side-effects were found.
- Indirect Prompt Injection (LOW): The framework is designed to ingest and process untrusted data (agent outputs) to reach consensus. While it acts as a data ingestion surface, it implements defensive logic and lacks capabilities for high-privilege side-effects. Ingestion points: The maker_red_flag and maker_vote tools ingest LLM outputs (SKILL.md). Boundary markers: Defined via constraints and schemas in the maker_generate_prompt tool (SKILL.md). Capability inventory: Operations are limited to internal logic, mathematical modeling, and prompt generation; no file system writes or command executions are defined. Sanitization: The framework implements a 'Red-Flagging' system to validate output format, length, and content before processing (technical_reference.md).
Audit Metadata