Risk Disclosure

Read this before building, testing, or exporting a strategy.

Stockey helps structure and test user-defined trading logic, but it does not remove market risk, provide investment advice, recommend trades, or guarantee any outcome.

Disclosure summary
read before use

Disclosure categories

Market risk
Backtest limitations
AI limitations
Broker/API risk
User-controlled runtime risk
User responsibility

Disclosure status

Trading risk
Always present
Backtest result
Not predictive
AI review
Not advice
Broker credentials
User controlled
Runtime use
User-controlled only
Stockey execution
Disabled / not provided

Responsibility boundary

User defines strategy
User selects data source
User reviews assumptions
User controls credentials
User controls runtime
User accepts live risk
No investment advice
No trade recommendations
No promised returns
Backtests are historical
AI is review-only
User-controlled environment

Plain-language summary

This page explains the main limitations and responsibilities when using Stockey.

Trading is risky

Markets can move against any strategy. Losses can exceed expectations, especially with leverage, derivatives, options, futures, intraday trading, gaps, and volatile instruments.

Backtests are limited

Historical tests show how a rule set behaved on available past data. They do not predict future performance.

AI is not an adviser

AI analysis may explain risks, assumptions, and test results, but it does not provide buy/sell calls, security recommendations, or portfolio advice.

Broker data varies

Historical data availability, instrument coverage, rate limits, and API behavior differ by broker, segment, instrument, and timeframe.

Exports are user-owned

Planned package exports are self-hosted and dry-run-first. The user controls credentials, environment variables, infrastructure, and any live package decisions.

Legal review

This disclosure should be reviewed by qualified counsel before legal reliance and may be updated as Stockey's product, policies, providers, and legal obligations evolve.

Stockey has a limited role.

Stockey may help with

  • Structuring user-defined strategy rules
  • Capturing risk assumptions in StrategySchema
  • Running historical tests with user-authorized data or CSVs
  • Showing data availability and data gaps
  • Producing AI-assisted review summaries
  • Preparing self-hosted Python package exports
  • Preparing broker and deployment checklists
  • Explaining package safety defaults

Stockey does not do

  • Provide investment advice
  • Recommend securities or trades
  • Sell trading calls
  • Guarantee profits or risk reduction
  • Execute live trades from Stockey SaaS
  • Store broker secrets permanently by default
  • Act as a broker or exchange
  • Manage portfolios or client money
  • Provide account-mirroring features
  • Automatically approve strategies for live use

Market and trading risks

Any trading strategy can lose money. Technical structure and testing cannot eliminate market risk.

Market movement

  • Price gaps
  • Volatility spikes
  • Liquidity shocks
  • Slippage
  • Spread widening
  • News events

Leveraged products

  • Futures risk
  • Options risk
  • Margin risk
  • Assignment/exercise complexity
  • Multi-leg strategy risk
  • Intraday leverage risk

Operational risk

  • Broker downtime
  • API failure
  • Incorrect instrument mapping
  • Network outage
  • Server misconfiguration
  • User credential errors
Risk controls can reduce ambiguity, but they cannot guarantee protection from losses.

Backtests are historical simulations, not forecasts.

Backtest Limitation Report

Diagnostic fields for historical backtest review.

Not predictive
Data source
broker historical data or CSV
Data range
limited by available historical data
Missing candles
reported where detected
Costs/slippage
assumptions only
Execution model
simplified compared with live markets
Liquidity
must be reviewed by user
Result status
diagnostic, not predictive
Past performance does not guarantee future results
Overfitting can make a strategy look better than it is
Data gaps can distort results
Transaction cost assumptions can materially change outcomes
Live execution can differ from historical testing
Broker-provided data may have limitations

AI analysis is review support, not trading advice.

What AI can do

  • Explain strategy assumptions
  • Summarize backtest behavior
  • Flag missing risk controls
  • Point out data limitations
  • Suggest further tests the user may choose to run

What AI cannot do

  • Recommend a stock, option, future, or trade
  • Tell the user when to enter or exit live markets
  • Guarantee accuracy
  • Guarantee returns
  • Replace professional judgement
  • Approve a strategy for live trading

Required user review

  • User reviews every assumption
  • User reviews every test result
  • User decides whether to export
  • User decides whether to run locally
  • User accepts responsibility
AI output may be incomplete, inaccurate, or unsuitable for a user's situation. It must be reviewed by the user.

Broker credentials and market data remain the responsibility of the user.

Stockey-side boundary

  • Historical data access only when user requests it
  • Session-only credential handling by default
  • No permanent broker-secret storage by default
  • No order placement from Stockey SaaS
  • Secrets excluded from generated ZIPs
  • .env.example values only

User-side responsibility

  • User enters broker credentials locally after export
  • User confirms broker permissions
  • User checks rate limits and data coverage
  • User controls static IP or VPS setup
  • User monitors API errors and logs
  • User accepts broker/API terms and operational risk
Broker APIs can change. Users must verify current broker capability, permissions, and limitations before relying on any exported package.

Self-hosted means user-controlled and user-responsible.

Stockey may generate package files and setup checklists, but the exported package belongs to the user.

Responsibility areaStockey prepares checklistUser controlsUser accepts responsibility
Local .env credentials
Broker login/session
Static IP setup
VPS/server security
Dry-run verification
Live enablement
Package monitoring
Logs and alerts
Broker/account risk
Final trading decision

Safety defaults

Dry-runenabled by default
Live modedisabled by default
Risk acknowledgementrequired before live use
Broker secretsnot stored permanently by default
Stockey SaaS ordersnot placed

Planned package exports start locked down.

Safety defaults

Dry-run mode enabled by default
Live mode disabled by default
Broker secrets excluded
Risk disclosure included
Risk acknowledgement required
Backtest report included where available
AI analysis marked as review-only
User must configure local environment
User must manually enable live mode outside Stockey
src/execution/live_guards.py
if DRY_RUN != false:
    block_live_orders()
if LIVE_TRADING_ENABLED != true:
    block_live_orders()
if RISK_ACKNOWLEDGED != true:
    block_live_orders()
if broker_credentials_missing:
    block_live_orders()

Before using Stockey, the user should understand:

I understand trading involves risk and possible loss.
I understand backtests do not predict future performance.
I understand AI analysis is not investment advice.
I understand Stockey does not provide trade recommendations.
I understand Stockey does not execute live orders from the SaaS.
I understand exported packages are self-hosted and user-controlled.
I understand I am responsible for broker credentials and user-controlled runtime operation.
I understand Stockey's terms, privacy policy, and disclosures may be updated over time.
Acknowledgement required in app flow
Read Terms and Privacy

No guarantees

Stockey does not guarantee:

Strategy profitability
Strategy accuracy
Backtest reliability
AI output correctness
Broker API availability
Data completeness
Order execution quality
Risk reduction
Live runtime success
Regulatory suitability for every use case

Use Stockey as an engineering and review tool, not as a guarantee of trading success.

Use Stockey with clear boundaries.

Define your rules, review your assumptions, understand the limits, and take responsibility before exporting a package or running it in a user-controlled environment.