The prompt is not an AST syntax analysis target but is separated for analysis intent and Retriever/LLM judgment direction
The AST Parser receives source code, file path, file type, original hash, and other source metadata
The Retriever should search the vector DB based on APIs, symbols, identifiers, and intent words extracted from the code fragment, not on file paths
The document written today needs to be reviewed again tomorrow, especially to ensure that the boundaries of AST fragments, direct chunks, Retriever queries, and trace metadata remain consistent throughout the document