Recipe catalog / deadline-risk

Grade the risk of missing a deadline

How at risk is the work of missing deadline given progress, from on track to already missed or impossible?

A supervisor or planner has a stated deadline and a progress report and needs a graded read on whether to escalate, replan, or leave the work alone.

Explore this recipe interactively · Source and implementation guide

Use deadline-risk in TypeScript

Install with npm install jev-recipes. Requires Node.js 22.9 or newer and ES modules. Set TYPESAFE_API_KEY in your server environment for live calls, which send input to TypeSafe and use API quota. See the installation guide.

import { deadlineRisk } from 'jev-recipes/deadline-risk';

const result = await deadlineRisk({
  "deadline": "The migrated customer records must be live in the new CRM by end of day Friday, per the signed contract. It is now Thursday at 6pm.",
  "progress": "The export from the old system finished Wednesday. The import script rejects about 30% of records with a schema mismatch that nobody has diagnosed yet. The engineer who wrote the field mapping is out until Monday. No fallback or partial go-live has been agreed with the customer.",
  "minConfidence": 0.8
});
console.log(result);

Input contract

{
  "input": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "deadline": {
        "type": "string"
      },
      "progress": {
        "type": "string"
      },
      "minConfidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      }
    },
    "required": [
      "deadline",
      "progress"
    ]
  },
  "result": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "model": {
        "type": "string"
      },
      "usage": {
        "type": "object",
        "properties": {
          "input_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          },
          "output_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          }
        },
        "required": [
          "input_tokens",
          "output_tokens"
        ],
        "additionalProperties": false
      },
      "status": {
        "type": "string",
        "enum": [
          "ready",
          "review"
        ]
      },
      "score": {
        "type": "number",
        "minimum": 0
      },
      "level": {
        "type": "integer",
        "minimum": 0,
        "maximum": 9007199254740991
      },
      "confidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      },
      "probabilities": {
        "type": "object",
        "propertyNames": {
          "type": "string"
        },
        "additionalProperties": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        }
      },
      "risk": {
        "type": "string",
        "enum": [
          "none",
          "minor",
          "moderate",
          "high",
          "missed"
        ]
      }
    },
    "required": [
      "model",
      "usage",
      "status",
      "score",
      "level",
      "confidence",
      "probabilities",
      "risk"
    ],
    "additionalProperties": false
  }
}

Saved example result

This hand-authored response demonstrates the contract. It is not a model accuracy measurement. Run it without an API key: npx jev-recipes demo deadline-risk.

{
  "model": "demo-fixture",
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0
  },
  "status": "ready",
  "score": 2.99,
  "level": 3,
  "confidence": 0.82,
  "probabilities": {
    "0": 0,
    "1": 0.01,
    "2": 0.08,
    "3": 0.82,
    "4": 0.09
  },
  "risk": "high"
}

Evaluation evidence

Use the evaluation guide to measure this decision on your own labeled cases.

Limitations

• Judges described progress against the described deadline. Date arithmetic, calendar lookups, and working-hour calculations belong in application code; state the time remaining in the text.
• Trusts the progress report as written. Optimistic or stale reports produce optimistic grades.
• A grade is not a plan. Escalation and replanning decisions belong to the caller.

Related recipes

• step-progress: Use step-progress to judge whether a single new observation moved the task forward, rather than how the whole effort stands against a deadline.
• task-complexity: Use task-complexity to grade how hard a task is before it starts, rather than how it is tracking once underway.