mcpField
P
Pine Script Strategy Audit

Pine Script Strategy Audit

Remote
github.com/tylerscomic-labFinance0Updated 13w ago
Repository

About Pine Script Strategy Audit

Pine Script Strategy Audit is an MCP server that statically checks Pine Script v6 strategies for patterns that can make live behavior differ from backtests. It returns the affected line and an explanation, helping strategy authors review code before deployment.

Key features

  1. 01Detection of order fill-timing traps
  2. 02Checks for ATR smoothing mismatches
  3. 03Detection of lookahead bias
  4. 04Checks for VWAP anchor drift
  5. 05Detection of exit behavior drift
  6. 06Pine v6 syntax checks

Use cases

  1. 01Reviewing a strategy before live trading
  2. 02Tracing a backtest/live behavior discrepancy

MCP support

It exposes the `audit_pine_strategy` tool to MCP clients. The tool accepts a Pine Script v6 strategy and reports detected issues with line references and explanations.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://pinescript-audit-mcp.mcpize.run/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

pinescript-audit-mcp

License: MIT Live on MCPize

An MCP server that statically audits Pine Script v6 strategies for the specific traps that compile clean, backtest beautifully, and then diverge from live behavior.

The problem this solves

Pine Script strategies can compile with zero errors and still behave differently live than in the backtester — fill-timing assumptions, ATR-smoothing mismatches, lookahead bias, VWAP-anchoring drift, and exit-drift are all silent, syntactically valid patterns that the compiler has no reason to flag. This catches them before they cost real money finding out live.

What it catches

  • Fill-timing traps — orders that backtest as filling at a price/bar the live broker can't actually match
  • ATR-smoothing mismatches — RMA vs. SMA vs. EMA smoothing assumptions that quietly change signal timing
  • Lookahead bias — request.security() and similar calls that leak future bar data into historical evaluation
  • VWAP-anchoring drift — anchor-period assumptions that behave differently on session boundaries live vs. in backtest
  • Exit-drift — exit logic that behaves differently depending on calc_on_every_tick and bar-close assumptions
  • Pine v6 syntax traps — v6-specific footguns that compile but change behavior from v5 equivalents

Tools

audit_pine_strategy

Runs the full static-analysis pass against a submitted Pine Script v6 strategy and returns every trap found, with the specific line and a plain-English explanation of the backtest-vs-live divergence it causes.

Use it

Hosted (recommended): MCPize — free tier, paid Pro tier for higher limits.

Self-host:

npm install
node server.js

Part of the AlgoForge suite

Prop-firm and quant-validation tools for algo traders: prop-rules-mcp, trade-journal-mcp, payout-calc-mcp, econ-calendar-mcp, montecarlo-validator-mcp, overfitting-audit-mcp, walkforward-validator-mcp, backtest-cost-sensitivity-mcp, pinescript-mcp.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.