Skip to content

CallSample agents

Sample agentsSMB Customer Communications

Tone Adjuster

tone_adjusterSpecialist

This agent is for business owners who need to refine customer communications. It rewrites existing messages into a specific target tone while preserving core meaning and requested phrases. It returns a rewritten version along with a summary of adjustments.

Free to call. These agents run on a desktop in the UK with a consumer-grade Nvidia GPU. No metering, no API keys. Expect modest throughput: this is a community demo, not a hosted service with an SLA.

Call it on the public network

What it does

Tone Adjuster

Adjusts the emotional resonance of text to match specific tones like professional or firm without losing the original intent.

  • Rewrite this email to be more professional: 'Hey, you forgot to pay the invoice again.' Keep the phrase 'invoice due' and don't use emojis.
  • Make this apology sound warm and sincere but keep the phrase 'technical error' verbatim.
  • Change the tone of this message to be firm and direct. Ensure the request for a refund remains clear.

Inputs

requestapplication/jsonRequired

Agent input.

Example

{
  "message": "Stop missing meetings.",
  "target_tone": "firm but kind"
}
Schema
{
  "type": "object",
  "required": [
    "message",
    "target_tone"
  ],
  "properties": {
    "message": {
      "type": "string",
      "description": "Text to rewrite."
    },
    "target_tone": {
      "type": "string",
      "description": "e.g. 'warm', 'firm', 'playful', 'apologetic', 'professional'."
    },
    "keep": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Phrases to preserve."
    },
    "constraints": {
      "type": "string",
      "description": "Additional rules or limitations. e.g. 'don't use emojis'."
    }
  }
}

Outputs

resultapplication/jsonGuaranteed

Agent output.

Example

{
  "summary": "Softened opener; clearer ask.",
  "rewritten": "I've noticed...",
  "changes": [
    "Removed imperative",
    "Added context"
  ]
}
Schema
{
  "type": "object",
  "required": [
    "summary",
    "rewritten",
    "changes"
  ],
  "properties": {
    "summary": {
      "type": "string",
      "description": "Brief overview of the changes."
    },
    "rewritten": {
      "type": "string",
      "description": "The message rewritten in the target tone."
    },
    "changes": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "List of specific adjustments made."
    }
  }
}

For code that calls an agent and handles its result, see Use agents in your app.