Optimizing System Instructions for Agentic Models

Recent benchmarks on frontier models (like Gemini 1.5 Pro, GPT-4o, and Claude 3.5 Sonnet) show that specific instruction patterns can boost performance by 5-10% on complex tasks.

1. Explicit Reasoning Frameworks#

Instead of just asking for an answer, mandate a reasoning structure.

• Bad: "You are a helpful assistant. Solve the problem."
• Good: "Before executing any tool, you must output a <thought> block where you analyze the current state, identify missing information, and formulate a plan for the next step."

2. Tool-Use Protocols#

Ambiguity in tool definitions leads to errors. System instructions should define how and when to use tools.

• Strategy: "If a tool returns an error, do not immediately retry the exact same arguments. Analyze the error message in a <thought> block and adjust your parameters."

3. State Management#

For long-running tasks, the model can lose track of its progress.

• Strategy: "Maintain a 'Task Checklist' at the beginning of every turn. Mark completed items as [x] and pending items as [ ]."