MIESC 6.4.0 - Complete Installation Guide¶

This guide covers the complete installation of MIESC from scratch, including all dependencies and tools.

Table of Contents¶

Quick Install
System Requirements
Step-by-Step Installation
Tool Installation by Layer
Docker Installation
Verification
Troubleshooting

Quick Install¶

For users who want to get started quickly with basic functionality:

# Clone repository
git clone https://github.com/fboiero/MIESC.git
cd MIESC

# Install MIESC and core dependencies (Slither is included)
pip install -e .

# Install the Solidity compilers Slither will need (picked per contract pragma)
pip install solc-select
solc-select install 0.4.26 0.5.17 0.6.12 0.7.6 0.8.20

# Verify installation
python scripts/verify_installation.py
miesc doctor

System Requirements¶

Minimum Requirements¶

Component	Requirement
OS	macOS 12+, Ubuntu 20.04+, Windows 10+ (WSL2)
Python	3.12 or higher
Memory	8 GB RAM
Disk	5 GB free space

Recommended Requirements¶

Component	Requirement
OS	macOS 14+, Ubuntu 22.04+
Python	3.12+
Memory	16 GB RAM
Disk	20 GB free space (for all tools)
Node.js	18+ (for Solhint)
Rust	1.70+ (for Aderyn, Foundry)
Go	1.21+ (for Medusa)

Step-by-Step Installation¶

Step 1: Clone Repository¶

git clone https://github.com/fboiero/MIESC.git
cd MIESC

Step 2: Create Virtual Environment (Recommended)¶

python3.12 -m venv .venv
source .venv/bin/activate  # Linux/macOS
# or: .venv\Scripts\activate  # Windows

Step 3: Install MIESC¶

# Basic installation (CLI + Slither)
pip install -e .

# Paper 1 reproducible profile: built-in ML detectors
pip install -e '.[ml]'

# Full installation (PDF, LLM clients, RAG, local APIs, ML)
pip install -e '.[full]'

# Development installation
pip install -e '.[dev]'

Extras: ml (scikit-learn detectors), pdf (WeasyPrint), mcp (MCP server, SDK 1.x pinned), llm (OpenAI/Anthropic clients), rag, django/api (local REST), full, dev.

Step 4: Install Solidity Compiler¶

pip install solc-select
# One version per major line you analyze. MIESC selects, per contract, the
# installed version that satisfies its pragma (SmartBugs-curated is mostly 0.4.x).
solc-select install 0.4.26 0.5.17 0.6.12 0.7.6 0.8.20
solc-select use 0.8.20

# Verify
solc --version

Step 5: Verify the Core Analysis Tools¶

Slither is a core dependency of MIESC, so it is already installed. Aderyn and Solhint complete the static layer used in the papers:

slither --version          # installed with MIESC
cargo install aderyn       # or: brew install aderyn
npm install -g solhint

Tool Installation by Layer¶

MIESC integrates 50 tools across 35 analysis modules, organized in 9 defense layers. Install tools based on your needs.

Layer 1: Static Analysis (3 tools)¶

Tool	Installation	Required
Slither	included in `pip install miesc`	Yes
Aderyn	`cargo install aderyn`	Recommended
Solhint	`npm install -g solhint`	Recommended

# Install the remaining Layer 1 tools (Slither ships with MIESC)
cargo install aderyn
npm install -g solhint

Layer 2: Dynamic Testing (4 tools)¶

Tool	Installation	Notes
Echidna	Binary from releases	Property-based fuzzer
Medusa	Binary from releases	Coverage-guided fuzzer
Foundry	`curl -L foundry.paradigm.xyz \\| bash && foundryup`	Complete toolkit
DogeFuzz	Built-in	No external install

# Foundry
curl -L https://foundry.paradigm.xyz | bash
foundryup

# Echidna (macOS)
brew install echidna

# Echidna (Linux) - download from GitHub releases
wget https://github.com/crytic/echidna/releases/latest/download/echidna-linux-x86_64.tar.gz
tar -xzf echidna-linux-x86_64.tar.gz
sudo mv echidna /usr/local/bin/

# Medusa - download from GitHub releases
# https://github.com/crytic/medusa/releases

Layer 3: Symbolic Execution (3 tools)¶

Tool	Installation	Notes
Mythril	`pip install mythril`	May conflict with slither
Manticore	`pip install manticore[native]`	Python 3.10 only
Halmos	`pip install halmos`	Foundry integration

# Mythril (install in separate venv if conflicts)
pip install mythril

# Halmos
pip install halmos

# Manticore (requires Python 3.10)
# Best installed in Docker or separate environment

Layer 4: Formal Verification (3 tools)¶

Tool	Installation	Notes
Certora	certora.com	Commercial, free tier
SMTChecker	Included with solc	Built-in
Wake	`pip install eth-wake`	Python framework

pip install eth-wake

Layer 5: AI Analysis (6 tools)¶

Tool	Installation	Notes
SmartLLM	Built-in + Ollama	Local LLM analysis
GPTScan	Built-in	Requires OpenAI API key
LLMSmartAudit	Built-in	Multi-agent LLM
GPTLens	Built-in	LLM auditor/critic
LlamaAudit	Built-in + Ollama	Llama-based analysis
iAudit	Built-in	Fine-tuned LLM auditor

# Install Ollama
curl -fsSL https://ollama.ai/install.sh | sh

# Pull required models
ollama pull qwen2.5-coder:14b   # default for code analysis (~9 GB)
ollama pull qwen2.5-coder:32b   # default_model for verification (~20 GB, 24+ GB RAM)

Layer 6: ML Detection (5 tools)¶

Tool	Installation	Notes
DA-GNN	Built-in	Graph neural network
SmartBugs-ML	Built-in	ML classifier
SmartBugs-Detector	Built-in	SWC-category detection
SmartGuard	Built-in	Pattern matching
Peculiar	Built-in	Crucial data-flow ML

All built-in, no external installation required.

Layer 7: Specialized Analysis (7 tools)¶

Tool	Installation	Notes
Threat Model	Built-in	Threat modeling
Gas Analyzer	Built-in	Gas optimization
MEV Detector	Built-in	MEV exposure
Clone Detector	Built-in	Contract clone detection
DeFi	Built-in	Flash loan, oracle
Advanced Detector	Built-in	Rug pull, honeypot
Upgradability Checker	Built-in	Proxy/upgradability risks

All built-in, no external installation required.

Layers 8–9 (Cross-Chain & ZK Security, Advanced AI Ensemble) are experimental modules on the multi-chain roadmap; the EVM core is Layers 1–7.

Layer 8: Cross-Chain & ZK Security (5 tools)¶

Tool	Installation	Notes
Cross-Chain	Built-in	Cross-chain risks
ZK Circuit	`cargo install circomspect`	ZK circuit analysis
Bridge Monitor	Built-in	Bridge security
L2 Validator	Built-in	L2-specific checks
Circom Analyzer	`npm install -g circom snarkjs`	Circom/Noir analysis

# For ZK circuit analysis
cargo install circomspect
npm install -g circom snarkjs

Layer 9: Advanced AI Ensemble (5 tools)¶

Tool	Installation	Notes
LLMBugScanner	Built-in + Ollama	deepseek-coder model
Audit Consensus	Built-in	Multi-LLM consensus
Exploit Synthesizer	Built-in	PoC exploit synthesis
Vuln Verifier	Built-in	Finding verification
Remediation Validator	Built-in	Fix validation

Docker Installation¶

For a complete, isolated environment with all tools pre-installed:

ARM64 (Apple Silicon)¶

docker build -t miesc:local .
docker run --rm -v $(pwd):/contracts miesc:local audit quick /contracts/MyContract.sol

x86_64 (Intel/AMD)¶

docker build --platform linux/amd64 -f Dockerfile.x86 -t miesc:local-x86 .
docker run --platform linux/amd64 --rm -v $(pwd):/contracts miesc:local-x86 audit quick /contracts/MyContract.sol

Pre-built Image¶

# Pull from GitHub Container Registry
docker pull ghcr.io/fboiero/miesc:latest    # Standard (multi-arch, ~2-3GB)
docker pull ghcr.io/fboiero/miesc:full      # Full - amd64 only (~8GB)

# Run audit
docker run --rm -v $(pwd):/contracts ghcr.io/fboiero/miesc:latest audit quick /contracts/MyContract.sol

# Or build locally (use the build script for ARM compatibility)
./scripts/build-images.sh standard
./scripts/build-images.sh full              # On ARM: prompts for native build
docker run --rm -v $(pwd):/contracts miesc:latest audit quick /contracts/MyContract.sol

Verification¶

After installation, verify everything is working:

# Run verification script
python scripts/verify_installation.py

# Or use the CLI
miesc doctor

Expected output:

MIESC v6.4.0 - Installation Verification
============================================================

1. Python Environment
----------------------------------------
  [OK] Python 3.12.x

2. Core Python Dependencies
----------------------------------------
  [OK] slither-analyzer (0.11.x)
  [OK] pydantic (2.x)
  ...

2b. Optional Python Extras
----------------------------------------
  [WARN] scikit-learn - optional, install with: pip install 'miesc[ml]'
  ...

5. MIESC Adapter Registration
----------------------------------------
  [OK] All 50/50 adapters registered
  [OK] 43/50 tools currently available (run `miesc doctor` for the list)

Installation Summary
============================================================
MIESC is ready to use!

Only Python and the core dependencies decide the exit code; missing optional tools are reported as [WARN]. The number of available tools depends on what you installed.

Troubleshooting¶

Mythril installation fails¶

Mythril may conflict with slither-analyzer due to z3-solver version requirements.

Solution: Install Mythril in a separate virtual environment or use Docker:

# Option 1: Separate venv
python3 -m venv mythril-env
source mythril-env/bin/activate
pip install mythril

# Option 2: Docker
docker run -it mythril/myth analyze /path/to/contract.sol

Manticore on ARM (Apple Silicon)¶

Manticore requires x86_64 architecture and Python 3.10.

Solution: Use Docker with emulation:

docker build --platform linux/amd64 -f Dockerfile.x86 -t miesc:x86 .

solc version issues¶

# Install the line the contract's pragma asks for; MIESC picks it automatically
solc-select install 0.4.26 0.5.17 0.6.12 0.7.6 0.8.20
solc-select use 0.8.20

Ollama not responding¶

# Start Ollama service
ollama serve &

# Check status
ollama list

# Pull the default models
ollama pull qwen2.5-coder:14b

Missing Rust/Cargo¶

# Install Rust
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
source ~/.cargo/env

Missing Node.js/npm¶

# macOS
brew install node

# Ubuntu
curl -fsSL https://deb.nodesource.com/setup_18.x | sudo -E bash -
sudo apt-get install -y nodejs

Permission denied errors¶

# Fix pip permissions
pip install --user -e .

# Or use virtual environment
python3 -m venv .venv
source .venv/bin/activate
pip install -e .

Next Steps¶

After installation:

Quick Start: miesc audit quick contract.sol
Full Audit: miesc audit full contract.sol
Local REST API: python -m miesc.api.rest --host 127.0.0.1 --port 8000
Static Reports: python -m miesc.utils.web_dashboard --results ./results --output ./dashboard
Reproduce the paper results: REPRODUCE.md
Documentation: fboiero.github.io/MIESC

Support¶

Issues: github.com/fboiero/MIESC/issues
Email: fboiero@frvm.utn.edu.ar
Documentation: fboiero.github.io/MIESC

Version: 6.4.0 | Last Updated: October 2026