Supported Languages

Vibe Analyzer supports AST parsing and LLM enrichment for 13 programming languages and file formats.

Full List

Note: Markdown is a special case. Headings, links, code blocks, and frontmatter metadata are extracted instead of programmatic constructs. This allows Markdown files to be used as a knowledge base: documentation, guidelines, standards, project legends. They are searchable via search_documentation and search_knowledge.

Extracted Element Categories

For All Programming Languages

Markdown Only

Element Support by Language

Multilingual Search Tags

Each extracted element receives tags in three languages. This allows searching for elements in Russian, English, or Chinese — regardless of the source code language.

Doc Comment Formats

Vibe Analyzer extracts documentation from specially formatted comments. Regular comments (//, #) are ignored.

AST Example

Source code (sample.py):

"""
Module documentation for Python testing
"""

import os
import sys
from datetime import datetime
from typing import List, Optional

# Regular comment - ignored


def add(a: int, b: int) -> int:
    """Adds two numbers"""
    return a + b


def multiply(a: int, b: int) -> int:
    """Multiplies two numbers"""
    return a * b


async def fetch_data(url: str) -> str:
    """Async function example"""
    return "data"


class User:
    """User class"""

    def __init__(self, name: str, age: int):
        """Constructor"""
        self.name = name
        self.age = age

    def get_name(self) -> str:
        """Get user name"""
        return self.name


class Config:
    """Config class"""
    debug: bool = False
    max_size: int = 1024


class Color:
    """Color enum (using class constants)"""
    RED = 1
    GREEN = 2
    BLUE = 3


MAX_SIZE: int = 1024
DEFAULT_TIMEOUT: int = 30

APP_NAME: str = "vibe-analyzer"

# Regular comment at the end - ignored

Extracted AST:

{
  "functions": [
    {
      "signature": "def add(a: int, b: int) -> int",
      "comments": ["Adds two numbers"]
    },
    {
      "signature": "def multiply(a: int, b: int) -> int",
      "comments": ["Multiplies two numbers"]
    },
    {
      "signature": "async def fetch_data(url: str) -> str",
      "comments": ["Async function example"]
    }
  ],
  "classes": [
    {
      "signature": "class User",
      "comments": ["User class"]
    },
    {
      "signature": "class Config",
      "comments": ["Config class"]
    },
    {
      "signature": "class Color",
      "comments": ["Color enum (using class constants)"]
    }
  ],
  "variables": [
    {
      "signature": "MAX_SIZE: int"
    },
    {
      "signature": "DEFAULT_TIMEOUT: int"
    },
    {
      "signature": "APP_NAME: str"
    }
  ],
  "imports": ["os", "sys", "datetime", "typing"],
  "header_comments": ["Module documentation for Python testing"],
  "tags": [
    "header_comments",
    "imports",
    "импорты",
    "导入",
    "variables",
    "переменные",
    "变量",
    "functions",
    "функции",
    "函数",
    "classes",
    "классы",
    "类"
  ]
}

Limitations

• Maximum file size for AST parsing: 10 MB (MAX_AST_FILE_SIZE constant)
• Default ignored directories: target, node_modules, __pycache__, .venv, venv, .git, .idea
• Default ignored files: .DS_Store, Thumbs.db, *.hprof, *.log
• Binary files: detected by extension, name, and content analysis, excluded from parsing
• Nested elements: methods inside classes are extracted as functions, variables inside functions/classes are not extracted