Evidence before guesses
Source calls, resolver links, known runtime behavior, and inference stay distinct. Incomplete results keep visible diagnostics.
PROJECT INTELLIGENCE · PUBLIC BETA
Turn MQL5 code into a map AI can reason about.
Junet.store shares practical AI, MQL5, and open-source engineering — starting with MQL5 CodeGraph, a local analysis tool for evidence-backed architecture, impact, and context tracing.
MITPython 3.11+v0.3.0Public beta
Evidence graph ready
Local-first, no API key required
The core analyzer, CLI, loopback dashboard, and local MCP runtime need no cloud account or model key.
Build in public, grow together
MIT source, open design discussions, reproducible bug reports, and visible contributor recognition.
FLAGSHIP PROJECT
MQL5 CodeGraph
Evidence-backed static code-graph intelligence for MQL5 repositories — local, deterministic, and usable from a CLI, dashboard, or AI agent.
MQL5 CodeGraph reads .mq5 and .mqh files in an operator-trusted repository, extracts structure and relationships, and produces a queryable JSON snapshot. Evidence origins stay explicit so an agent cannot quietly promote inference into fact.
01
Extract includes, classes, structs, enums, functions, methods, event handlers, and calls.
02
Query symbols, context, impact, directed paths, and GraphML exports.
03
Use a loopback dashboard for search, graph navigation, and source evidence.
ONE-COMMAND INSTALL
Start local. Keep source on your machine.
Install the public v0.3.0 wheel, then select a trusted local MQL5 repository to analyze.
python -m pip install "mql5-codegraph[reference,mcp] @ https://github.com/junet03/mql5-codegraph/releases/download/v0.3.0/mql5_codegraph-0.3.0-py3-none-any.whl"The core analyzer needs no model API key. Python 3.11+ is required.
KNOWLEDGE HUB
Start with practical knowledge
Five foundation articles, fully written in Vietnamese and English.
Use AI to read MQL5 architecture, trace impact, and review changes
A practical workflow that starts from an entry point, gathers bounded context, checks blast radius, and separates static evidence from compiler and runtime truth.
ReviewImpactMQL5
Read article What is MQL5 CodeGraph, and why does AI need evidence-backed graphs?
AI can read many lines of code and still lose its bearings. A code graph connects symbols, calls, and event handlers to explicit evidence.
MQL5CodeGraphAI agents
Read article Connect MQL5 CodeGraph MCP to Codex and other AI agents
Understand the local MCP boundary, register mql5-codegraph-mcp per client, select a separate trusted project root, and maintain honest RAM-only snapshots.
MCPCodexAI agents
Read article