Fix ETAs

Get predicted fix timelines for all open CVEs.

population and issuesConsidered say what the predictions were computed over, so the promise in the line above is checkable from the response rather than taken on trust. Closed nixpkgs security issues are excluded: a predicted fix date for work already finished is not a prediction, and for a period this endpoint published thousands of them. Estimates are based on multiple signals including severity, upstream fix availability, assignee status, linked PRs, package health, issue age, and comment activity.

Endpoint

GET /api/v1/fix-etas

Response

{
  "success": true,
  "data": {
    "predictions": {
      "CVE-2025-1234": {
        "estimatedDays": 14,
        "confidence": 55,
        "etaDate": "2026-02-24",
        "range": {
          "low": "2026-02-17",
          "high": "2026-03-10"
        },
        "factors": [
          {
            "signal": "high_severity",
            "impact": "accelerator",
            "detail": "High severity increases urgency"
          },
          {
            "signal": "upstream_fix_available",
            "impact": "strong_accelerator",
            "detail": "Upstream fix exists - version bump likely"
          }
        ],
        "package": "openssl",
        "issueNumber": 42
      }
    },
    "population": "open_tracker_issues",
    "issuesConsidered": 562,
    "totalPredicted": 25,
    "summary": {
      "within7Days": 5,
      "within30Days": 15,
      "within90Days": 22,
      "beyond90Days": 3
    },
    "lastUpdated": "2026-02-10T12:00:00.000Z"
  },
  "timestamp": "2026-02-10T12:00:00.000Z"
}

Response Fields

Per-CVE Prediction

Field	Type	Description
estimatedDays	number	Predicted days until fix (1-180)
confidence	number	Confidence percentage (10-90%)
etaDate	string	Predicted fix date (ISO date)
range.low	string	Optimistic estimate date
range.high	string	Pessimistic estimate date
factors	array	Signals that influenced the prediction
package	string	Affected package name
issueNumber	number	GitHub issue number

Prediction Factors

Signal	Impact	Description
critical_severity	strong_accelerator	Critical CVEs get prioritized (0.4x time)
high_severity	accelerator	High severity increases urgency (0.6x)
upstream_fix_available	strong_accelerator	Upstream fix exists (0.4x)
assigned	accelerator	Issue has an assignee (0.5x)
linked_pr	strong_accelerator	PR already exists (capped at 7 days)
orphaned_package	strong_decelerator	No maintainer (2x time)
active_maintainers	accelerator	Multiple maintainers (0.8x)
stale_issue	decelerator	Open > 60 days (1.3x)
active_discussion	accelerator	5+ comments (0.8x)

Summary

Field	Type	Description
within7Days	number	CVEs predicted to be fixed within 7 days
within30Days	number	CVEs predicted to be fixed within 30 days
within90Days	number	CVEs predicted to be fixed within 90 days
beyond90Days	number	CVEs predicted to take longer than 90 days

Example

curl https://api.vulnpatch.dev/api/v1/fix-etas

Code Examples

JavaScript

async function getUpcomingFixes() {
  const response = await fetch('https://api.vulnpatch.dev/api/v1/fix-etas');
  const { data } = await response.json();

  // CVEs likely to be fixed soon
  const soonFixes = Object.entries(data.predictions)
    .filter(([, p]) => p.estimatedDays <= 7 && p.confidence >= 50)
    .sort((a, b) => a[1].estimatedDays - b[1].estimatedDays);

  console.log(`Expected fixes this week: ${soonFixes.length}`);
  soonFixes.forEach(([cve, p]) => {
    console.log(`  ${cve} (${p.package}): ~${p.estimatedDays}d, ${p.confidence}% confidence`);
  });
}

Python

import requests

response = requests.get("https://api.vulnpatch.dev/api/v1/fix-etas")
data = response.json()["data"]

print(f"Summary: {data['summary']}")

# Stuck CVEs (high estimate, low confidence)
stuck = {cve: p for cve, p in data["predictions"].items()
         if p["estimatedDays"] > 90 and p["confidence"] < 30}

for cve, p in stuck.items():
    factors = ", ".join(f["signal"] for f in p["factors"] if "decelerator" in f["impact"])
    print(f"  {cve}: ~{p['estimatedDays']}d - blockers: {factors}")

Use Cases

• Sprint planning: Know which CVEs are likely to be fixed soon
• Escalation: Identify CVEs with long ETAs that may need intervention
• Reporting: Provide stakeholders with estimated remediation timelines
• Prioritization: Focus on CVEs where a fix is imminent vs. stalled

Caching

Predictions are computed every 30 minutes via cron.

Related Endpoints

• GET /api/v1/package-health - Package health (input to ETA model)
• GET /api/v1/analytics/time-to-fix - Historical time-to-fix benchmarks
• GET /api/v1/rebuild-impact - Rebuild blast radius