Hi!

Yes, Agents are capable of utilizing multiple tools.

For your specific situation, I recommend refining the AGENT SYSTEM prompt. While your prompt outlines the agent's objectives and available tools, it lacks instructions on how to execute them. A common best practice is to include a STEP-BY-STEP or PROCESS section within your prompt to detail the agent's workflow comprehensively.

Consider also incorporating examples for both INPUT and the expected OUTPUT.

Additionally, define the desired OUTPUT in a dedicated section.

Hope this proves helpful. All the best!

Can anyone share a "hardcore" system message template? (if you come across one)

Thanks!

Certainly,

You can begin with this template and utilize the parts you require:

System Prompt Template

Context

You are a [ROLE/PERSONA] tasked with [PRIMARY PURPOSE]. Your expertise lies in [DOMAIN/FIELD], and you are equipped with [CAPABILITIES/TOOLS]. Your interactions should maintain a [TONE/STYLE] and be directed towards [TARGET AUDIENCE].

Goal

To achieve [PRIMARY OBJECTIVE]. Specifically:

• [Secondary objective 1]
• [Secondary objective 2]
• [Secondary objective 3]

Input Information Definition

{
  "input_type": "string/object/array",
  "required_fields": {
    "field_1": "description of field 1",
    "field_2": "description of field 2",
    "field_3": "description of field 3"
  },
  "optional_fields": {
    "optional_1": "description of optional field",
    "optional_2": "description of optional field"
  },
  "constraints": {
    "max_length": "number",
    "format": "expected format",
    "validation_rules": ["rule 1", "rule 2"]
  }
}

Process

1. Initial Analysis: Begin by reading and parsing the input to comprehend [WHAT TO UNDERSTAND].
2. Validation: Verify that the input adheres to the specified format and constraints.

3. Processing: Proceed with [MAIN PROCESSING STEPS]:

   • [Step 1 description]
   • [Step 2 description]
   • [Step 3 description]

4. Quality Check: Confirm that the output satisfies all requirements.

5. Final Output: Structure the response according to the defined output format.

Rules

• Do not [PROHIBITION 1].
• Always [REQUIREMENT 1].
• Never [PROHIBITION 2].
• Must [REQUIREMENT 2].
• If [CONDITION], then execute [ACTION].
• When encountering [SCENARIO], respond with [RESPONSE].
• Prioritize [PRIORITY 1] over [PRIORITY 2].
• Handle errors by employing [ERROR HANDLING APPROACH].

Output Format

{
  "status": "success/error",
  "result": {
    "primary_output": "main result",
    "metadata": {
      "confidence_score": "0.0-1.0",
      "processing_time": "timestamp",
      "version": "version_number"
    },
    "additional_info": {
      "field_1": "value",
      "field_2": "value"
    }
  },
  "errors": ["error message if any"],
  "warnings": ["warning message if any"]
}

Examples

Example 1: [SCENARIO NAME]

Input:

{
  "example_field": "example_value"
}

Expected Output:

{
  "status": "success",
  "result": {
    "primary_output": "expected result"
  }
}

Example 2: [ERROR SCENARIO]

Input:

{
  "invalid_field": "invalid_value"
}

Expected Output:

{
  "status": "error",
  "errors": ["Description of what went wrong"]
}

Additional Notes

• [SPECIAL CONSIDERATION 1]
• [SPECIAL CONSIDERATION 2]
• For edge cases involving [SCENARIO], follow [HANDLING INSTRUCTION].
• Remember to [IMPORTANT REMINDER].