- Add PLATFORM query type that bypasses RAG and uses a lighter model
- Introduce OLLAMA_MODEL_NAME_CONVERSATIONAL env var to route CONVERSATIONAL
and PLATFORM queries to a separate (smaller) Ollama model
- Replace raw message history in classifier with compact intent history
(classify_history) to eliminate anchoring bias in small models
- Add <history_rule> and <platform_priority_rule> to classifier prompt so
the model evaluates each message independently while still resolving
ambiguous references from prior turns
- Add fast-path detection for known platform-injected prompt prefixes
- Add PLATFORM_PROMPT for account/metrics/usage responses
- Persist classify_history in classify_history_store alongside session_store
- Document decisions in ADR-0008
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Introduced sections on Persistence, Connectors, and Native ORM, detailing the avapConnector, ORM commands, and data access abstraction.
- Documented System Utilities and Transformation, covering time management, string manipulation, and security operations.
- Explained Function Architecture and Scopes, including function definition, invocation, and middleware usage.
- Provided a Master Example that integrates various sections to demonstrate practical application.
- Detailed the dynamic nature of AVAP™ as a programming language, including dynamic typing and memory management.
- Established notation conventions and lexical analysis processes for code clarity and structure.
- Outlined data types and structures available in AVAP™, emphasizing their usage in program development.
- Discussed variable management, including local and global variables, and best practices for comments.
- Explained expressions in AVAP™, including types, operators, and practical examples with lists.
rafa-ruiz
acano