Monte-Neo strategy verifier
Independent checker for AI-written trading backtests: look-ahead, costs and overfitting
- Pricing
- Open source
- Platforms
- CLI, Linux, macOS
- Runs
- Locally
- Version
- 0.54.0
- Listing updated
- 2 days ago
About Monte-Neo strategy verifier
Monte-Neo strategy verifier is an open-source MCP server for finance and payments. It runs locally on your machine and is started by your AI client.
Monte-Neo examines the backtest rather than the trading idea. Given price data and either the strategy code or its positions, it returns one of four verdicts (PASS, PASS_WITH_WARNINGS, NEEDS_MORE_EVIDENCE or REJECT), the checks behind it, next steps for an agent to fix, and a reproducible strategy-verdict certificate that can be signed.
Checks fall into four families. Look-ahead probes truncate data and perturb the future and also run a static lint with 23 rules. Economics checks net return after costs, break-even cost in basis points and execution delays. Statistics include probabilistic and Deflated Sharpe priced by the number of trials, a Monte Carlo timing test and walk-forward. Integrity checks catch broken OHLCV data.
Newer releases add a repair mode that minimises defects instead of maximising return, alpha validation, a validation dossier mapped to SR 11-7, PRA SS1/23 and MiFID II RTS 6, and a quote-arrival-time check for millisecond look-ahead. It installs from PyPI, runs as a CLI, a GitHub Action or an MCP server for Claude Code, Codex, Gemini CLI and Cursor.
Install
Run in a terminal
claude mcp add monte-neo -- uvx monte-neoAdd to claude_desktop_config.json
{
"mcpServers": {
"monte-neo": {
"command": "uvx",
"args": [
"monte-neo"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"monte-neo": {
"command": "uvx",
"args": [
"monte-neo"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"monte-neo": {
"type": "stdio",
"command": "uvx",
"args": [
"monte-neo"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"monte-neo": {
"command": "uvx",
"args": [
"monte-neo"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"monte-neo": {
"command": "uvx",
"args": [
"monte-neo"
]
}
}
}What it does
- Look-ahead probes
- Truncation, future-perturbation and a 23-rule static lint.
- Cost realism
- Net return after fees, break-even bps and execution delay.
- Overfitting statistics
- Deflated Sharpe priced by number of trials, plus walk-forward.
- Signed certificates
- Reproducible verdicts that others can re-check.
- Repair mode
- Fixes defects without optimising for return.
- CI integration
- A GitHub Action fails the pull request on REJECT.
Good for
- Review a backtest an AI agent just wrote before trusting its Sharpe
- Fail a pull request when a strategy leaks future data
- Screen strategy submissions at a prop firm or course
- Re-check someone else's signed certificate
- Find the latency at which a quote-based edge vanishes
Pricing
Open source
Install from PyPI with pip
Prices change. Check neozork.github.io before you buy.
Monte-Neo strategy verifier FAQ
What verdicts does it give?
PASS, PASS_WITH_WARNINGS, NEEDS_MORE_EVIDENCE and REJECT, with CLI exit codes 0, 0, 1 and 2.
How do I install it?
pip install monte-neo, then run monte-neo verify with your OHLCV file and strategy.
Does it judge the strategy idea?
No, it checks the backtest rather than the idea.
Which agents can use it?
The site names a Claude Code plugin and the MCP server for Codex, Gemini CLI and Cursor.
Is Monte-Neo strategy verifier free?
Yes. Monte-Neo strategy verifier is open source, so you can use and self-host it for free.
How do I add Monte-Neo strategy verifier to Claude Code?
Run this command in your terminal: claude mcp add monte-neo -- uvx monte-neo
What are the best alternatives to Monte-Neo strategy verifier?
The closest alternatives to Monte-Neo strategy verifier are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Monte-Neo strategy verifier

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools