Can I submit Pine Script or custom strategy code?

No. The tool accepts only six versioned preset IDs and never executes caller-provided code or arbitrary expressions.

Is the holdout used to pick the winning variant?

No. Selection uses development data only; holdout results are reported afterward and cannot change the selected historical candidate.

Can AI predict stock prices with this MCP tool?

No. AI cannot predict stock prices with certainty, and this MCP tool does not promise future returns. It gives an agent bounded data or validation results that still require independent review, risk controls, and source checking.

Agent workflows

Trading Idea Validation MCP Tool for AI Agents

validate-trading-idea runs one exact US company or ADR through one of six fixed long-only daily presets and three fixed variants. Signals are observed at the close, executed at the next valid open with 5 bps friction per side, selected only on the 70% development sample, and reported separately on the 30% holdout sample.

Connect Pineify MCP View all MCP tools

MCP tool name

validate-trading-idea

Example prompt

Use Pineify validate-trading-idea for AAPL with rsi-mean-reversion over 5y and compare all variants.

Direct answer

What validate-trading-idea does

The tool accepts preset IDs rather than code or a strategy DSL. Dividend-adjusted OHLCV, fixed warmup, minimum-trade rules, stable tie-breaking, and no-lookahead timing are versioned.

Historical candidate, sensitivity, benchmarks, and holdout results remain visible even when the selected variant is ineligible or underperforms.

Contract

What the agent sends and receives

Inputs

`symbol`: one exact active US company or ADR.
`preset`: one of six fixed strategy IDs.
`lookback`: 3y or 5y; optional `detail`: compact or standard.

Structured output

`trading-idea-validation-v1` structured content.
Exactly three sensitivity variants, one historical candidate, development/holdout metrics, and up to four regime slices.
Data coverage, adjusted-history counts, benchmarks, warnings, and methodology.

Capabilities

Where this MCP tool fits

Run fixed presets

Supports RSI mean reversion, SMA crossover, EMA pullback, MACD trend, Bollinger mean reversion, and price breakout.

Protect holdout isolation

Selects the candidate using development metrics only, then reports the untouched chronological holdout.

Report comparable metrics

Returns trades, return, volatility, Sharpe, drawdown, exposure, and separate underlying/SPY benchmarks.

Agent workflow

A bounded call from question to review

1
Choose a fixed hypothesis
Select a supported preset instead of sending executable code.
2
Review development eligibility
Check minimum trades and the deterministic candidate-selection record.
3
Judge holdout evidence separately
Compare candidate, underlying, and SPY results without reselecting on holdout.

Prompt examples

Questions an agent can route to this tool

"Use Pineify validate-trading-idea for AAPL with rsi-mean-reversion over 5y and compare all variants."

"Validate the MSFT sma-crossover preset over 3y and explain development versus holdout performance."

Operator notes

How I review the result

I call the result a historical candidate, not a best or future parameter set.

I keep no-trade and minimum-trade failures visible instead of optimizing them away.

Boundaries to keep in the prompt

Only six fixed long-only daily presets and three fixed variants are supported.
Historical backtests are sensitive to sample, costs, and market regime and do not predict future performance.
The tool does not execute user code, place trades, or provide personalized advice.

Pineify MCP is an information and code-validation tool, not investment advice. It does not promise returns, place trades, or modify a portfolio. Review timestamps, source fields, code diagnostics, and risk assumptions before acting.

MCP tool family