Skip to content
Gauge

ax-check.com

Checkly

checklyhq.com

ax-check.com

Checked 24 September 2026

A

100/100

Agents need your login

Checkly's docs and pricing are fully machine-readable, no gates.

All 23 checklist items passed: quickstart, CLI, API, MCP and skills all resolve, and pricing tiers (Hobby $0, Starter $29, Team $80) are stated with clear limits.

1 fix to start with

Concrete changes, backed by what we found.

Onboarding

Give agents a clear login handoff and a way to resume setup.

Agents cannot progress through onboarding without human help. Provide an agent-accessible path through authentication and setup.

Recorded coding sessions

Same task, 3 models, no checklyhq.com account. Recorded 24 September 2026 on Pi.

3 recorded · 1m 32s

1m 32s · recorded

Session findings

Select a finding to see it in context.

piv0.84.2
modelQwen 3.8 Maxcwd/sandbox/repo
Help me build a simple example using Checkly. Tell me how pricing works, and briefly tell me whether this product will be easy for you to manage. Let me know if you get blocked. If this product has no developer workflow you can act on, say so plainly and stop. Stay light: use the hosted product through its SDK or API. Do not start local service stacks or wait for long-running commands; if the quickstart requires either, say so plainly and stop.
Thinking…

The user is asking for an example using Checkly (a monitoring-as-code product). Let me check the environment — is the repo empty? Let me verify whether there's internet access, npm, etc.

Checkly pricing: Checkly has a free plan, then paid plans based on the number of checks/API calls. Let me check access to the environment.

$ ls -la /sandbox/repo; node --version 2>/dev/null; npm --version 2>/dev/null{"content":[{"text":"total 16\ndrwxr-xr-x 3 root root 4096 Sep 18 08:00 .\ndrwxr-xr-x 7 root root 4096 Sep 24 05:42 ..\ndrwxr-xr-x 8 root root 4096 Sep 18 08:00…ctrl+o
{"content":[{"text":"total 16\ndrwxr-xr-x 3 root root 4096 Sep 18 08:00 .\ndrwxr-xr-x 7 root root 4096 Sep 24 05:42 ..\ndrwxr-xr-x 8 root root 4096 Sep 18 08:00 .git\n-rw-r--r-- 1 root root  408 Sep 18 08:00 .gitignore\nv22.23.2\n10.9.8\n","type":"text"}]}
$ curl -s -m 10 -o /dev/null -w "%{http_code}" https://api.checklyhq.com 2>&1; echo; curl -s -m 10 -o /dev/null -w "%{http_code}" https://registry.npmjs.org/checkly 2>&1{"content":[{"text":"200\n200","type":"text"}]}
Thinking…

Network access is working. Let me scaffold a minimal Checkly project with the checkly npm package (the CLI / Monitoring-as-Code). This requires a CHECKLY_API_KEY and an account ID — without them, I can't deploy. I can build the project structure locally (this isn't a long-running service, just files and npm install), but deploying requires API credentials, which I don't have. That's a blocker for the hosted part. I should report it.

Should I quickly fetch the pricing info from their site? Checkly's pricing: there's a free plan, then paid plans based on the number of API/browser checks (monthly minutes/usage). Let me try fetching the pricing page.

Network is up and npm works. Let me check Checkly's pricing info and set up a minimal project.

$ curl -s -m 15 https://www.checklyhq.com/pricing/ | sed 's/<[^>]*>/ /g' | tr -s ' \n' ' \n' | grep -iA2 -B2 '\$\|free\|month' | head -80{"content":[{"text":"(function () {\n window.dataLayer = window.dataLayer || [];\n--\n }\n})();\n :root{--font-inter:'Inter', 'Inter Fallback'} Live webinar: In…ctrl+o
{"content":[{"text":"(function () {\n window.dataLayer = window.dataLayer || [];\n--\n }\n})();\n :root{--font-inter:'Inter', 'Inter Fallback'} Live webinar: Inside Yext’s SLA monitoring. Oct 1 · 12 PM ET · 6 PM CEST. Register now → Checkly - Home Product Detect Uptime Monitoring Measure the availability of your digital footprint Heartbeat Monitoring Catch cron jobs and backups that fail silently Synthetic Monitoring Simulate real user interactions across your stack Testing Catch issues before production with an AI-powered test reporter Communicate Status Pages Communicate app availability to your customers Alerts Contextual alerting to notify the team right away Resolve AI Root Cause Analysis Automated root cause analysis powered by AI agents Traces Powerful OTel tracing for deeper insights Getting Started Start for free Book a demo Talk to Sales Monitoring as Code Solutions Personas Developers Checks in TypeScript, in your repo, deployed with your app SRE &amp; Platform The synthetic layer for your observability stack QA Engineers Run your Playwright suite as production monitors Engineering Managers Give every team ownership of its own monitors Use Cases Critical user flows Watch login, checkout, and signup from real browsers API &amp; backend monitoring Validate endpoints, chains, and auth on a schedule Tests to production monitors Promote tagged specs from CI to scheduled checks Observability consolidation Replace the synthetic module bolted onto your APM Reliability for AI-generated code Verify what your agents ship before your users do Industries E-commerce Protect checkout and revenue paths around the clock Financial services Prove uptime and latency against strict SLAs SaaS &amp; B2B software Keep every tenant&#x27;s core workflows verified Moving from another tool? vs. Datadog vs. New Relic vs. Pingdom vs. Catchpoint Developers Developers Documentation Learn key concepts and features API Docs Build on the Checkly REST API CLI Docs Bring monitoring to your terminal Quickstart Set up your first check fast Guides In-depth Playwright &amp; OTel guides MCP Server Connect Checkly to your AI tools Reference Agent Skills Terraform Pulumi All Integrations Changelog Resources Featured Checkly and the Agentic Software Layer Read article → Resources Blog Read about the latest news at Checkly Learn Tips and best practices for learning Playwright and more Webinars Register or view webinars on-demand Events Meet the team in person Community Public Roadmap See Checkly&#x27;s active customer outcomes and product direction Community Slack Connect with the Checkly Community Customers Pricing Login Start for free Open Navigation Product Solutions Developers Resources Customers Pricing Start for free Login \n html:has(.pricing-page-overflow-guard) #nav [inert] {\n display: none;\n--\n .pricing-card-stagger { animation-duration: 0.01ms !important; animation-delay: 0ms !important; }\n }\n Reliability starts with a plan Choose yours. Then scale it up. Detect Communicate Resolve Test &amp; Monitor your APIs and websites with comprehensive Uptime and Synthetic monitoring. Monthly Annual Save up to 20 % Hobby Perfect for personal projects and learning $ 0 per month Get Started Uptime Monitors 10 Maximum Frequency 2 Min. Locations 6 Retries Synthetic Monitoring Browser/Playwright Checks 1,000 API Checks 10,000 Round Robin Locations 6 Private Locations — Plus Access To: Email, Slack, Webhooks for Alerting Starter Great for startups and growing projects $ 24 per month Try for Free Uptime Monitors 50 Maximum Frequency 1 Min. Locations 6 Retries Synthetic Monitoring Browser/Playwright Checks 3k API Checks 25k Round Robin Locations 6 Private Locations — Plus Access To: Email, Slack, SMS, Webhooks for Alerting Team We Recommend Ideal for growing teams and businesses $ 64 per month Try for Free Uptime Monitors 75 Maximum Frequency 30 Sec. Locations 22 Retries Synthetic Monitoring Browser/Playwright Checks 12k API Checks 100k Round Robin Locations 22 Private Locations Plus Access To: All Alerting Channels Enterprise Custom solutions for large organizations Custom Contact Sales Uptime Monitors Custom Maximum Frequency 1 Sec. Locations 22 Retries Synthetic Monitoring RR &amp; Parallel Scheduling Locations 22 Private Locations Plus Access To: All Alerting Channels World-class engineering and SRE teams depend on Checkly to deliver reliable digital experiences Plan Details Swipe left to see all plans Select your billing cycle Monthly Annually Detect Monitor your APIs and websites with comprehensive uptime and synthetic monitoring Hobby $ 0 per month, billed annually Get Started Starter $ 24 per month, billed annually Try for Free Team $ 64 per month, billed annually Try for Free Enterprise Get a custom quote Contact Sales Uptime Monitoring Hobby Starter Team Enterprise Protocol Support The network protocols available for monitoring your endpoints and services. HTTPS, TCP, DNS, ICMP, SSL/TLS, gRPC, Traceroute, Heartbeat HTTPS, TCP, DNS, ICMP, SSL/TLS, gRPC, Traceroute, Heartbeat HTTPS, TCP, DNS, ICMP, SSL/TLS, gRPC, Traceroute, Heartbeat HTTPS, TCP, DNS, ICMP, SSL/TLS, gRPC, Traceroute, Heartbeat Uptime Monitors The number of uptime monitors you can create to track the availability of your websites, APIs, and services. 10 50 75 Custom Additional Monitors (25) Purchase additional monitors in blocks of 25 to scale beyond the included amount. — $8 $8 Custom Scheduling How checks are distributed across locations. Round Robin rotates through locations sequentially. Parallel runs from all locations simultaneously. Round Robin Round Robin Round Robin Round Robin &amp; Parallel Public Locations Global data center locations from which your checks are executed. More locations help detect regional outages. 6 N. Virginia, N. California, Frankfurt, London, Singapore, Sydney Max 3 locations per check 6 N. Virginia, N. California, Frankfurt, London, Singapore, Sydney Max 3 locations per check All 22 Locations All 22 Locations Private Locations Run checks from within your own infrastructure to monitor internal services not exposed to the public internet. — — Single Retry Automatically retry a failing check once before triggering an alert, reducing false positives from transient issues. Automatic Retries Automatically retry a failing check multiple times before triggering an alert. — — Synthetic Monitoring Hobby Starter Team Enterprise Browser Checks Simulate real user interactions using a headless browser to validate critical user flows like logins, checkouts, and form submissions. Playwright Check Suites Run full Playwright test suites as scheduled monitors. Group multiple related browser checks into a single suite. Browser Check / Playwright Check Suite Runs The total number of browser check or Playwright check suite executions included per month. 1,000 per month 3,000 per month 12,000 per month Custom Additional Browser Check / Playwright Check Suite Runs Purchase additional browser check runs beyond your included amount at a per-1k rate. — $4.00 per 1k $4.00 per 1k Custom Browser Check / Playwright Check Suite Overages (1k) The per-1k cost charged automatically when you exceed your purchased browser check runs. — $6.50 $6.25 Custom API &amp; Multistep Checks Monitor API endpoints with HTTP requests. Multistep checks chain multiple API calls together to test complex workflows. API Check Runs The total number of API check executions included per month. 10,000 per month 25,000 per month 100,000 per month Custom Additional API Check Runs Purchase additional API check runs beyond your included amount at a per-10k rate. — $1.80 per 10k $1.80 per 10k Custom API Check Overages (10k) The per-10k cost charged automatically when you exceed your purchased API check runs. — $2.60 $2.50 Custom Automatic Retries Automatically retry a failing check before triggering an alert, reducing false positives from transient issues. Scheduling How checks are distributed across locations. Round Robin rotates through locations sequentially. Parallel runs from all locations simultaneously. Round Robin &amp; Parallel Round Robin &amp; Parallel Round Robin &amp; Parallel Round Robin &amp; Parallel Public Locations Global data center locations from which your checks are executed. More locations help detect regional outages. 6 N. Virginia, N. California, Frankfurt, London, Singapore, Sydney Max 3 locations per check 6 N. Virginia, N. California, Frankfurt, London, Singapore, Sydney Max 3 locations per check All 22 Locations All 22 Locations Private Locations Run checks from within your own infrastructure to monitor internal services not exposed to the public internet. — — SSL certificate expiry alerts API and browser checks alert you before the TLS certificate on their target domain expires. For dedicated certificate monitoring, create an SSL monitor. 10 Second Frequency Run checks as frequently as every 10 seconds for near real-time monitoring of critical endpoints. — — Visual Regression Testing Automatically capture and compare screenshots to detect unintended visual changes on your website. — — Alerting Hobby Starter Team Enterprise Email Alerts Receive alert notifications via email when a check fails or recovers. Slack Alerts Send alert notifications directly to your Slack channels. Webhook Alerts Send alert payloads to any HTTP endpoint for custom integrations with your existing tools. Incident.io, Rootly &amp; more Integrate with incident management platforms like Incident.io, Rootly, PagerDuty, OpsGenie, and more. — — SMS Alerts Receive alert notifications via text message. The number shown is your included monthly SMS credits. 0 100/month 200/month Custom Phone Alerts Receive alert notifications via automated phone calls. The number shown is your included monthly call credits. 0 0 200/month Custom Maintenance Windows Schedule maintenance windows to pause alerting during planned downtime, preventing false positive alerts. — — Communicate Keep your team and users informed with fast, beautiful status pages and dashboards. Hobby $ 0 per month, billed annually Get Started Starter $ 9 per month, billed annually Try for Free Team $ 30 per month, billed annually Try for Free Enterprise Get a custom quote Contact Sales Status Pages Hobby Starter Team Enterprise Services The number of individual services or components you can display on your status page. 20 25 50 100 Subscribers The number of people who can subscribe to receive status page updates via email or RSS. 250 500 1000 2000 Custom Domain Host your status page on your own domain (e.g., status.yourcompany.com) instead of a Checkly subdomain. — Custom CSS Apply your own CSS styles to fully customize the look and feel of your status page. — — Password Protection Restrict access to your status page with a password, useful for internal-only status pages. — — White Labeled Remove all Checkly branding from your status page for a fully branded experience. — — — Incident Automation Create, update, and resolve incidents directly from your status page to keep users informed during outages. — — Maintenance Windows Schedule maintenance windows on your status page so subscribers are informed of planned downtime in advance. — — Dashboards Hobby Starter Team Enterprise Dashboards Custom dashboards to visualize and share your monitoring data with your team or stakeholders. 1 1 10 Custom Resolve Diagnose issues faster with AI-powered root cause analysis Hobby $ 0 per month, billed annually Get Started Starter $ 12 per month, billed annually Try for Free Team $ 39 per month, billed annually Try for Free Enterprise Get a custom quote Contact Sales AI Root Cause Analysis Hobby Starter Team Enterprise Invocations The number of AI-powered root cause analysis requests you can run per month to diagnose check failures. 10 50 150 Custom Automated RCA Automatically trigger root cause analysis when a check fails, so you get insights without manual intervention. Platform Developer tools, support, and enterprise-grade security features Hobby $ 0 per month, billed annually Get Started Starter $ 24 per month, billed annually Try for Free Team $ 64 per month, billed annually Try for Free Enterprise Get a custom quote Contact Sales General Hobby Starter Team Enterprise Users The number of team members who can access your Checkly account. 1 3 10 Custom Data Hobby Starter Team Enterprise Raw data retention How long individual check run results, logs, and traces are stored before being purged. 7 days 7 days 30 days 180 days Aggregated data retention How long aggregated metrics like response times and availability percentages are retained for trend analysis. 30 days 30 days 1 year 25 months Integrations Hobby Starter Team Enterprise MCP Connect MCP clients to Checkly to inspect checks, test sessions, RCA, result assets, status pages, incidents, and account context. Checkly CLI Manage checks, run tests, and deploy monitoring as code from your terminal using the Checkly CLI. Terraform provider Define and manage your Checkly monitoring infrastructure as code using HashiCorp Terraform. Pulumi Provider Define and manage your Checkly monitoring infrastructure as code using Pulumi. Prometheus Endpoint Export your monitoring metrics to Prometheus for integration with Grafana and other observability tools. — — Customer Support Hobby Starter Team Enterprise Premium Support Priority access to the Checkly support team with faster response times and dedicated attention. — — — Onboarding Support Guided onboarding sessions to help your team get set up and running with Checkly quickly. — — — Dedicated Customer Success Engineer A dedicated Checkly engineer assigned to your account to help optimize your monitoring strategy. — — — 24x7 Phone Escalation Around-the-clock phone escalation support for critical issues that need immediate attention. — — — Contracts &amp; Invoicing Custom contracts and invoicing options tailored to your organization&#x27;s procurement needs. — — — Security Review Access to Checkly&#x27;s security documentation and processes for your organization&#x27;s vendor assessment. — — — Security &amp; Privacy Hobby Starter Team Enterprise SOC 2 Type II Compliant Checkly is SOC 2 Type II certified, ensuring your data is handled with the highest security standards. Multi-Factor Auth Add an extra layer of security to your account with multi-factor authentication (MFA). SAML/SSO Enable single sign-on (SSO) via SAML to manage access through your identity provider (e.g., Okta, Azure AD). — — — 99.9% Uptime SLA A guaranteed 99.9% uptime service level agreement for the Checkly platform itself. — — — Client Certificates Use client-side TLS certificates to authenticate checks against services that require mutual TLS (mTLS). — — — Service API Keys Create non-user-bound API keys for service-to-service integrations with the Checkly API. — — — Frequently Asked Questions Transform your development workflow with a code-based monitoring process that&#x27;s programmable, fast, and reliable. Platform General Billing Platform What is an Uptime Monitor? Uptime Monitors are simple monitors that measure the availability and performance of your services. There are eight types: URL, TCP, DNS, ICMP, SSL, gRPC, and Traceroute monitors, plus Heartbeats from passive processes like Cron Jobs. What is an API Check? API Checks are monitors that do POST, GET, and other typical HTTP requests. You have full control over the payload, headers, etc. You can also create assertions to check if the response from your API is valid and performant. They can be scheduled to run as often as every 10 seconds, while a 1-second interval is available upon request. What is a Browser Check? Browser Checks are monitors based on Node.js scripts using the open-source Playwright testing framework. Each Browser Check uses an actual browser to execute navigation, fill out forms and validate if these interactions have the correct results. This is also commonly called “synthetic monitoring”. For billing, we count every executed script as a check run. Browser check runs have a timeout threshold of 4 minutes. What is a Playwright Check Suite? Playwright Check Suites are similar to Browser Checks, but are able to run your Playwright test suites completely natively, with custom dependencies, shared storage state, and more. For billing, Playwright Check Suites are equal to a single Browser Check running for 30 seconds. For each 30 seconds they run, it will consume one check run. i.e., a 45 second Playwright Check Suite will consume 2 runs. What is a Status Page? Status pages provide transparent communication about your service health and incidents to customers, users, and stakeholders. Checkly’s status pages automatically reflect your monitoring results and create a professional interface for incident communication. Learn more about Status Pages . Are Traces in any plan? On our Enterprise plan, these are available as add-ons and can be purchased separately. Traces are not officially available on our other plans yet. If you&#x27;d like early access, please reach out to support . Learn more about Traces . How can I add features like private locations or visual regression testing? These are available in Team and Enterprise plans. What is the difference between Round Robin and Parallel scheduling? Round Robin scheduling runs checks sequentially across locations. Parallel allows them to run simultaneously for faster feedback. General How does the Hobby plan work? By signing up for the Hobby plan, you get access to almost all of Checkly’s features. You do not need a credit card to get started. You get access to 10 Uptime Monitors and can execute up to 10,000 API and 1,000 Browser Check runs per month. Can I upgrade or downgrade my plan? Yes. You can change between our Hobby, Starter, and Team plans using our self-serve billing portal. For upgrades, you&#x27;ll be charged the pro-rated amount for the remainder of your current billing cycle (whether monthly or annual). You&#x27;ll get immediate access to all the new features and benefits of your upgraded plan. For example, if you upgrade halfway through your monthly cycle, you&#x27;ll only pay for the remaining 15 days at the new plan rate. When you downgrade your plan, the change will be scheduled to take effect at the end of your current billing cycle. This means you can continue to enjoy all the features of your current plan until your next billing date. Your next invoice will reflect the lower plan rate and your account will be downgraded to the new plan features. For upgrades or downgrades involving our Enterprise plan, please contact sales . Do you support early startups or nonprofits? Yes! We do support early-stage startups that are younger than two years by providing a 30% discount on all plans. Nonprofits qualify for a 50% discount. Apply here Billing How do you bill? Prepaid plans, like the Starter and Team plans, are paid upfront at the beginning of the subscription cycle. Overages will be added to the invoice of the next billing cycle. How are uptime monitors billed? Uptime monitors (URL, TCP, DNS, ICMP, SSL, gRPC, Traceroute, Heartbeat) are billed based on the number of monitors in your account. Once you reach the limit, you won&#x27;t be able to create more unless you purchase additional uptime monitors. How are synthetic checks billed? Synthetic checks are billed based on check runs. We have two categories: API &amp; Multistep Check runs API Checks: Each execution counts as one run. Multistep Checks: Each request counts as one run. For example, if your Multistep Check script makes 3 requests, then that counts as 3 runs each time your check executes. Browser Check &amp; Playwright Check Suite runs Browser Check: Each execution counts as one run. Playwright Check Suite: For every 30 seconds of execution time, they consume one check run. i.e., a 45 second Playwright Check Suite will consume 2 runs. Each plan includes a specific number of check runs: The Hobby plan is hard-capped at 1…
Thinking…

Got the pricing info. Next, build a minimal Checkly project. The Checkly CLI (checkly npm package) supports monitoring-as-code: checkly init, checkly deploy (requires CHECKLY_API_KEY + account ID). No local services needed. Installing the checkly package is fine (quick).

But deploying requires an API key from a Checkly account, which I don't have on hand. This is a blocker that should be reported. Still, I can create a sample project locally and validate it with checkly test? checkly test runs checks locally—actually it uses a local runtime for API checks, and for browser checks it needs a local Playwright installation… checkly test for API checks doesn't require deploying to hosted; it runs locally. That might be quick and require no account. Let me try installing the checkly CLI and creating a simple API check.

Plan:

  1. npm install checkly
  2. Create checkly.config.ts and checks/api.check.ts (or .spec.ts)
  3. Try npx checkly test (runs locally, quick)—this validates the developer workflow without an account.
  4. Note that deployment requires CHECKLY_API_KEY—an account is needed, and that's the blocker to report.

Let's proceed lightly. Install checkly.

Pricing info gathered. Now let me scaffold a minimal Checkly "Monitoring as Code" project using their CLI/SDK — no local services needed.

$ cd /sandbox/repo && npm init -y >/dev/null && npm install checkly --no-audit --no-fund 2>&1 | tail -3{"content":[{"text":"\nadded 253 packages in 27s\n","type":"text"}]}
Thinking…

Alright, now let's create the project files: checkly.config.ts and a simple API check. Checkly MaC: checks are placed in checks/*.check.ts and use the ApiCheck and Frequency constructors from 'checkly/constructs'. Example of a config file:

import { defineConfig } from 'checkly';

export default defineConfig({
  projectName: 'Checkly Example',
  logicalId: 'checkly-example',
  checks: {
    locations: ['us-east-1', 'eu-west-1'],
    checkMatch: '__checks__/*.check.ts',
    runtimeId: '2024.09',
  },
});

Check file:

import { ApiCheck, Frequency } from 'checkly/constructs';

new ApiCheck('api-health-check', {
  name: 'Public API health check',
  frequency: Frequency.EVERY_1M? 

Frequency options: EVERY_5M and the like are strings like '5 minutes'… the Frequency enum: EVERY_10S, EVERY_30S, EVERY_1M, EVERY_5M, EVERY_10M, etc. To save costs, use EVERY_5M.

const check = new ApiCheck('api-health-check-1', {
  name: 'JSONPlaceholder API',
  frequency: Frequency.EVERY_5M,
  locations: ['us-east-1', 'eu-west-1'],
  request: {
    url: 'https://jsonplaceholder.typicode.com/posts/1',
    method: 'GET',
    assertions: [
      AssertionBuilder.status().equals(200),
      AssertionBuilder.jsonBody('$', 'id').equals(1),
    ],
  },
});

AssertionBuilder might be imported from 'checkly/constructs'. Let's create the files, and then run npx checkly test to run them locally (fast, no account needed). checkly test runs against the local runtime — it downloads the runtime… actually checkly test executes the checks locally using a bundled runtime; does the first run download a Docker image? No — checkly test has historically executed locally in-process. The newer CLI's checkly test runs on a local runtime that doesn't require Docker. Anyway, let's just try it with a timeout.

Now creating the project files:

write /sandbox/repo/checkly.config.ts{"content":[{"text":"Successfully wrote 460 bytes to /sandbox/repo/checkly.config.ts","type":"text"}]}
write /sandbox/repo/__checks__/posts-api.check.ts{"content":[{"text":"Successfully wrote 888 bytes to /sandbox/repo/__checks__/posts-api.check.ts","type":"text"}]}
Thinking…

Alright, let's verify: run npx checkly test locally (no account needed, and it should be quick). I'll set a timeout just in case it tries to download the runtime.

Now let me validate the checks run correctly (the CLI can run API checks locally without an account):

$ cd /sandbox/repo && timeout 120 npx checkly test 2>&1 | tail -20{"content":[{"text":" Error: Run `npx checkly login` or set `CHECKLY_API_KEY` & \n `CHECKLY_ACCOUNT_ID` in your environment or .env file.\n","type":"text"…ctrl+o
{"content":[{"text":"    Error: Run `npx checkly login` or set `CHECKLY_API_KEY` & \n    `CHECKLY_ACCOUNT_ID` in your environment or .env file.\n","type":"text"}]}
Thinking…

Even local testing requires authentication credentials (because it uses the hosted test runtime). In other words, I'm blocked: no account/API key. Let me quickly check the CLI help to confirm the workflow, and if possible verify the config parsing without credentials: npx checkly deploy --help, or alternatively there might be a way to validate the config. Probably npx checkly --help. There's no local-only test mode without an account (the newer CLI uses cloud runtimes). So the remaining steps require a Checkly account + API key — I should make it clear that's the blocking factor.

As a final sanity check, let me verify that the TypeScript imports resolve (at the compile level) — e.g., with npx tsc --noEmit? That would require installing typescript and checkly's type definitions. Quick check: does 'checkly/constructs' exist as an export? Let me check the exports in node_modules/checkly/package.json.

$ cd /sandbox/repo && node -e "const p=require('./node_modules/checkly/package.json'); console.log(p.version); console.log(JSON.stringify(p.exports||p.main))" && node -e "const c=require('checkly/constr…{"content":[{"text":"9.5.0\n{\".\":\"./dist/index.js\",\"./constructs\":\"./dist/constructs/index.js\",\"./loader\":\"./dist/loader/index.js\",\"./util\":\"./di…ctrl+o
cd /sandbox/repo && node -e "const p=require('./node_modules/checkly/package.json'); console.log(p.version); console.log(JSON.stringify(p.exports||p.main))" && node -e "const c=require('checkly/constructs'); console.log('ApiCheck:', typeof c.ApiCheck, '| Frequency.EVERY_5M:', c.Frequency.EVERY_5M, '| AssertionBuilder:', typeof c.AssertionBuilder)"

{"content":[{"text":"9.5.0\n{\".\":\"./dist/index.js\",\"./constructs\":\"./dist/constructs/index.js\",\"./loader\":\"./dist/loader/index.js\",\"./util\":\"./dist/util/index.js\"}\nApiCheck: function | Frequency.EVERY_5M: Frequency { frequency: 5, frequencyOffset: undefined } | AssertionBuilder: function\n","type":"text"}]}
Thinking…

Imports have been resolved. Deployment/testing is blocked due to credential issues. I'll wrap things up with a summary, pricing info, notes on the admin, and blockers.

Done. Here's the summary:

What I built

A minimal Checkly Monitoring-as-Code project in /sandbox/repo (no local services involved — it's all config that deploys to the hosted product):

  • checkly.config.ts — project config (locations: us-east-1, eu-west-1; tags; check discovery pattern)
  • __checks__/posts-api.check.ts — an ApiCheck (via the checkly/constructs SDK) that hits a public REST endpoint every 5 minutes and asserts status 200, response time < 5s, and JSON body fields
  • package.json with checkly CLI v9.5.0 installed

I verified the SDK constructs resolve and are valid (ApiCheck, Frequency.EVERY_5M, AssertionBuilder all load correctly).

How pricing works (verified from checklyhq.com/pricing today)

  • Hobby — $0/mo (no credit card): 10 uptime monitors, 10k API check runs and 1k browser check runs per month (hard-capped — checks stop when exhausted). Email/Slack/webhook alerts.
  • Starter — $24/mo: 50 monitors, 3k browser / 25k API runs included. Extra runs $4/1k browser, $1.80/10k API; overages $6.50/1k and $2.60/10k. SMS alerts.
  • Team — $64/mo: 75 monitors, 12k browser / 100k API runs, all 22 locations, private locations, phone alerts.
  • Enterprise — custom.
  • Billing is per check run, not per check: frequency, parallel locations, and retries each multiply usage. Playwright suites consume 1 run per 30s of execution. Annual billing saves up to 20%.
  • My example check (every 5 min, round-robin across 2 locations) ≈ 288 runs/day ≈ 8.6k runs/mo — fits in the free Hobby tier.

Will it be easy for me to manage?

Yes. Checkly is unusually agent-friendly: everything is TypeScript in the repo, deployed via npx checkly deploy (or Terraform/Pulumi), checks are diffable code, and they even publish an MCP server and agent skills specifically for AI tooling. The whole lifecycle — create, test, deploy, destroy — is CLI-driven with no local infrastructure to babysit.

Blocked (one step remaining)

Actually deploying/running requires a Checkly account. npx checkly test and npx checkly deploy both fail without credentials — even the "local" test run uses their hosted runtime. To unblock:

  1. Sign up free at app.checklyhq.com/signup (no credit card for Hobby)
  2. Create an API key in account settings, then:
    export CHECKLY_API_KEY=... CHECKLY_ACCOUNT_ID=...
    npx checkly test    # dry-run the check against the hosted runtime
    npx checkly deploy  # ship it as a live scheduled monitor
    

Give me those two env vars and I can finish the deploy in one step.

AX Fundamentals

23 checks. Open an area for details.

This check is a quick preview. Test more tasks and models with Gauge Agents.

Test your product with Gauge Agents

AX Check by Gauge