Configuration

Display Configuration

You can control where the plugin appears in the cTrader interface from the Plugin Menu. The AI Trading Agent can be displayed in:

• Trade Watch
• Active Symbol Panel (ASP)

This allows you to place the agent where it best fits your trading workflow.

AI Configuration

AI Model

Select the AI model used by the Trading Agent. Different models may vary in speed, reasoning depth, and response quality.

API Key

To use ChatGPT-powered AI models, you must provide your own API key.

• The key is stored only on your device.
• All requests are sent directly to ChatGPT.
• We do not have access to your API key and it is never shared with any third party.

Reasoning Level

Choose the AI’s reasoning depth. Higher reasoning levels improve context awareness and decision quality but may increase response time and token usage.

Temperature

Controls how creative or deterministic the AI responses are.

• Lower values produce more consistent and conservative outputs.
• Higher values allow more exploratory and adaptive responses.

Max Tokens

Sets the maximum number of tokens the AI can use per response. This helps control response length, cost, and performance.

System Prompt

Edit the main system prompt that defines the core behavior of the AI Trading Agent. This prompt shapes how the agent thinks, responds, and interacts with market data and your commands.

Default AI Settings

AI Model

Default: gpt-5-mini (or equivalent reasoning-capable model)

Why: Strong market reasoning, stable responses, and good prompt interpretation.

Reasoning Level

Default: Medium

Why: Provides solid contextual understanding without excessive latency or token usage.

Use Higher When: Analyzing complex market conditions or multi-asset strategies.

Use Lower When: Scalping or requiring faster responses.

Temperature

Default: 0.3 – 0.4

Why: Keeps responses consistent, disciplined, and less speculative — ideal for trading decisions.

Lower Values: More deterministic, rule-focused behavior.

Higher Values: More creative exploration (not recommended for live trading).

Max Tokens

Default: 2000 – 3000

Why:

• Enough for clear reasoning and explanations without unnecessary verbosity or cost.
• Preserves recent conversation context
• Allows proper reasoning and trade validation
• Prevents truncated or shallow responses

Increase When: Requesting detailed analysis or strategy breakdowns.

Decrease When: Focusing on quick execution or low-latency prompts.

For prompt trading, the AI is usually doing:

• Interpreting intent
• Checking market context
• Producing a clear action + short reasoning

That typically fits well within 2000–2500 tokens. Setting the limit at 2000–2500 gives enough headroom without encouraging over-analysis or unnecessary verbosity.

In trading, short, decisive responses are a feature, not a limitation.

When it can feel too low

Increase max tokens if you are:

• Asking for deep multi-timeframe analysis
• Requesting full strategy breakdowns
• Running post-trade reviews or explanations
• Combining multiple assets in one prompt

Practical presets (history-aware)

• Live trading with chat history:2000 – 2500 → balanced and responsive
• Strategy refinement / multi-step prompts:3000 – 4000 → deeper context and continuity
• Research / learning sessions:4000+ → long-form reasoning and reviews

Best practice for prompt trading

• Keep only relevant recent messages in history
• Trim confirmations or repeated system messages
• Increase max tokens as session complexity grows

System Prompt

Default: Use the provided system prompt

Why: It is optimized for trading context, risk awareness, and prompt interpretation.

Modify Carefully: Changes directly affect agent behavior and decision style.

API Key

Default: User-provided

Security Note:

• Stored locally on your device only
• Never logged or shared
• Requests go directly to ChatGPT

Recommended Presets

• Scalping / Fast Execution
• Reasoning: Low–Medium
• Temperature: 0.2–0.3
• Max Tokens: 4900–6000
• Swing / Position Trading
• Reasoning: Medium–High
• Temperature: 0.3–0.4
• Max Tokens: 10000–15000