- Home
- Skills
- Testing & Debugging
- User Swarm — Staging Journey Testing
Works with the AI Agents you already use
User Swarm — Staging Journey Testing
Simulate hundreds of users on staging and compare their actions with dashboards, emails, SMS, push notifications, scheduled jobs, and test payments.
$19
User Swarm — Staging Journey Testing
Example session with this skill installed
Run the included demo shop with 200 simulated users on staging and compare dashboard revenue and abandoned-cart emails against the user ledger.
- Read your context and instructions
- Compiled the user swarm
Illustrative report: included demo shop
This example explains the two intentional bugs documented in the bundled demo. It is not a claim about results from a live customer app.
| Check | Expected | Observed behaviour in the demo | Result |
|---|---|---|---|
| Abandoned-cart emails | Every abandoned cart receives its email | At 200 users, 5 of 30 abandoned carts are missed | FAIL |
| Dashboard net revenue | Revenue accounts for refunds | Dashboard revenue omits refunds | FAIL |
What the agent investigates
- Compare users who abandoned carts with recipients recorded by the notification catcher.
- Inspect the scheduled abandoned-cart job for a batch limit.
- Compare the action ledger and refund records with dashboard revenue calculations.
Deliverables
The configured run produces HTML and Markdown reports, a CSV action ledger, expected-versus-actual checks, and evidence for missing recipients. Actual results depend on the app and the configured checks.
Next steps
Review the reported discrepancies, approve any app fixes separately, and rerun the same project configuration. Cleanup starts with a preview and runs only after approval.
Connects securely to your tools. The creator never sees your data.
About this skill
Find the bugs that appear when real user journeys run together. User Swarm helps an AI coding agent configure realistic personas, simulate concurrent activity on staging, and compare a shared action ledger with what the application reports and sends.
What it checks
- Completed, abandoned, dropped-off, and cancelled or refunded journeys.
- Dashboard and database totals against recorded user activity.
- Email, SMS, and push delivery through test notification catchers.
- Delayed reminders and abandoned-cart jobs using project-specific fast-forward hooks.
- Test payments and cleanup of tagged swarm data.
Included
A reusable Node.js engine, built-in notification catcher, Mailpit configuration, HTML and Markdown reports, a project template, a demo shop with two intentional bugs, reference guides, and an evaluation fixture.
How it works
Your agent first confirms the staging target, user journeys, behaviour mix, and checks. It creates a project-specific config, runs 1 user, then 10, then the full swarm, and reports expected versus actual results. Cleanup is previewed and requires approval.
Requirements and boundaries
Designed for Claude Code, with plain-text instructions and Node.js scripts. Requires Node.js 18+ and access to your staging app and code. Each app needs its own personas, checks, and integration hooks. Optional browser journeys require Playwright. Production testing and live payment keys are blocked. Notification routing or app changes require your approval. This is a configurable testing workflow, not a zero-setup test service.
Cannot run against production environments. Browser-based testing requires Playwright installation. Not designed for unit tests or simple API benchmarks.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
Requires