BAML
A DSL for building production ready applications powered by LLMs
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for BAML, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
BAML is an open-source framework designed to make reliable, production ready LLM agents. It's extremely fast (built in rust), and works exactly the same with every language (python, typescript, ruby, java, etc). BAML supports every LLM and makes every model better at structured outputs by using its SAP (schema-aligned parsing) technique.
Key features and capabilities
- structured outputs for every model
- Support for every model (openai, anthropic, gemini, bedrock, vllm, ..)
- support for every programming language (python, typescript, ruby, java, go, ..)
- Lifecycle management (development, deployment, monitoring)
Use cases
- Parse bank statements
- Build multi-agent workflows
- chatbots
- virtually anything