API Breaking-Change Radar — never let an API update break your production

Get alerted before a public API breaks your integration. This actor watches the APIs you depend on, detects breaking changes as they happen — via OpenAPI spec diffs and GitHub release/RSS classification — and pushes a structured alert to your webhook (Slack, or any endpoint). Built for engineering teams, platform developers and AI agents that must not be surprised by upstream changes.

Use this tool when…

• Your production depends on third-party APIs (Stripe, GitHub, Twilio, OpenAI, Discord, Slack…) and a silent breaking change would cost you an incident.
• You want a scheduled watchdog that diffs OpenAPI specs and classifies releases as breaking / non-breaking.
• Your AI agent or SRE bot needs a structured breaking-change feed — via the Apify MCP server, agents can call it with zero glue code.
• You want evidence, not vibes: every alert includes the exact diff that triggered it.

Features

• OpenAPI diff engine (oasdiff): snapshots the spec, hashes it, and diffs against the previous run — detects removed endpoints, changed types, renamed fields, removed enum values.
• GitHub release & RSS classification: rule-based classification of release notes ("Breaking" sections prioritized to kill false positives).
• State persistence in a key-value store: only NEW changes alert, never the same one twice.
• Webhook alerts (Slack-compatible JSON) with severity levels, including a critical-alert tier.
• Dry-run mode: full detection pipeline with zero billing events — free evaluation and QA.
• Batch watch list: monitor multiple APIs per run with a safety cap.
• AI-agent ready: README and input schema readable by LLMs through MCP.

Use cases

Production incident prevention

Run it daily on the APIs your product calls. Get the Slack alert the day a field disappears — not when customers complain.

Dependency hygiene for platform teams

Track the upstream surface you actually depend on, with a history of what changed and when.

Release monitoring for developer tools

Watch the frameworks and services your stack relies on and plan migrations before the deadline.

Agentic SRE workflows

Let an agent summarize "what broke this week across our dependencies" from the structured event feed.

Input

Field	Type	Required	Description	Example
apis	array	yes	APIs to watch. Type openapi: url must point to a raw OpenAPI spec. Type github: repo releases.	[{"type":"openapi","name":"Stripe","url":"https://.../openapi.json"}]
webhookUrl	string	no	POST endpoint called at the end of each run with a JSON summary	"https://hooks.slack.com/..."
dryRun	boolean	no	Full detection without charging events	true
maxApis	integer	no	Safety cap on APIs processed per run	10

{
  "apis": [
    {"type": "openapi", "name": "Stripe", "url": "https://example.com/openapi.json"},
    {"type": "github", "name": "oasdiff", "repo": "oasdiff/oasdiff"}
  ],
  "webhookUrl": "https://hooks.slack.com/services/XXX",
  "dryRun": true
}

Output

Per detected change: api, changeType (breaking | non-breaking), severity, summary, evidence (the diff or release reference), detectedAt. A run summary record with counts per API.

Pricing (pay per event)

Event	Price
run-started	$0.50
api-checked	$0.005 per API checked
breaking-change-detected	$0.50 per breaking change found
critical-alert	$1.00 per critical alert sent

No subscription. Platform margin: Apify takes 20%. dryRun: true runs the full pipeline with zero charges — evaluate for free. Tiered Apify plans (Bronze/Silver/Gold) get progressively lower prices.

Agentic payments (x402 / USDC)

This actor supports x402 agentic payments: AI agents without an Apify account can call it and pay per call in USDC on Base — verified live.

FAQ

Which APIs can it watch? Any public OpenAPI spec URL, and any public GitHub repository (releases). Six major APIs have diffable public specs today: Stripe, GitHub, Twilio, OpenAI, Discord, Slack.

How do I run it daily? Schedule it on Apify (cron UI) — set dryRun for free monitoring, or live mode for full alerting with your webhook.

False positives? The classifier prioritizes explicit "Breaking" sections in release notes and oasdiff's own breaking-level rules — tested against noisy release feeds.

Does it work with Claude, Cursor, GPT or n8n? Yes, via the Apify MCP server.

More tools from this toolbox

• RAG Web Browser
• Review Radar
• Full toolbox: apify.com/travelmonitorlab

---

Built and maintained by Adaga solutions — full-stack & AI agents in production, Luxembourg.