Churn Risk Detector

churn_risk_detectorSpecialist

This agent monitors customer communications to identify signals of cancellation intent. It analyzes message content alongside account context to return a risk assessment and a specific retention strategy.

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

Churn Risk Detector

Detects churn indicators in transcripts and suggests calibrated recovery plays based on customer tenure and contract value.

Analyze this support transcript and tell me if the customer is at risk of churning.
Scan this email for cancellation intent and suggest a retention play for a high-ARR account.
Review this customer message and flag any immediate signals that require human escalation.

Inputs

requestapplication/jsonRequired

Agent input.

Example

{
  "message": "We're considering moving to a competitor.",
  "account_context": "tenure_months: 24, arr_usd: 60000"
}

Schema

{
  "type": "object",
  "properties": {
    "message": {
      "type": "string",
      "description": "Customer message or transcript to analyze."
    },
    "account_context": {
      "type": "string",
      "description": "Tenure, ARR, last activity, etc."
    }
  },
  "required": [
    "message"
  ]
}

Outputs

resultapplication/jsonGuaranteed

Agent output.

Example

{
  "summary": "High risk: explicit competitor mention.",
  "risk_level": "high",
  "signals": [
    "Mentioned competitor",
    "24mo tenure"
  ],
  "retention_play": {
    "recommended_action": "CSM call within 24h",
    "urgency": "high",
    "owner_type": "CSM",
    "talking_points": [
      "Recent product investments",
      "Loyalty discount option"
    ]
  },
  "escalate": true
}

Schema

{
  "type": "object",
  "required": [
    "summary",
    "risk_level",
    "signals",
    "retention_play"
  ],
  "properties": {
    "summary": {
      "type": "string",
      "description": "Brief overview of risk and suggested action."
    },
    "risk_level": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high",
        "imminent"
      ],
      "description": "Severity of cancellation risk. e.g. high."
    },
    "signals": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Identified indicators of cancellation intent."
    },
    "retention_play": {
      "type": "object",
      "properties": {
        "recommended_action": {
          "type": "string"
        },
        "urgency": {
          "type": "string"
        },
        "owner_type": {
          "type": "string",
          "description": "e.g. CSM, Sales, Exec"
        },
        "talking_points": {
          "type": "array",
          "items": {
            "type": "string"
          }
        }
      },
      "description": "Suggested strategy to prevent cancellation."
    },
    "escalate": {
      "type": "boolean",
      "description": "Whether to trigger immediate human intervention."
    }
  }
}

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