Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsUAT Test Plan for Chatbots, Shopify + Workflows: Go/No-Go
Unlike generic "write test cases" prompts: named happy paths, edge/failure cases, data checks and a human go/no-go list — never fakes pass results.
$4
UAT Test Plan for Chatbots, Shopify + Workflows: Go/No-Go
Example session with this skill installed
Generate a UAT pack for a Make.com automation that takes a Typeform lead, checks for a duplicate in HubSpot, and either updates the contact or creates a new one before alerting a Slack channel.
- Read your context and instructions
- Compiled the uat test plan
I have generated the UAT pack for the HubSpot and Slack lead automation. This includes happy paths for new and existing contacts, along with edge cases for API failures and duplicate entries.
Automation QA pack: Typeform to HubSpot/Slack Lead Flow
Environment: Not specified — confirm staging vs production
Status: Draft for human execution — results not pre-filled
1. Scope (evidence-based)
- Workflow under test: Typeform submission to HubSpot CRM with Slack notification.
- In scope: Duplicate checking logic, HubSpot contact creation/update, Slack channel alerts.
- Out of scope: Typeform UI styling, HubSpot email marketing sequences post-creation.
- Unknowns: Slack workspace permissions, HubSpot API rate limits.
2. Actors & surfaces
| Actor | Surface | Role in flow |
|---|---|---|
| Lead | Typeform | Submits contact information |
| HubSpot API | Backend | Manages contact records |
| Slack Bot | Channel | Posts lead notifications |
| Admin | HubSpot | Verifies data integrity |
3. Critical happy paths
| ID | Flow | Preconditions | Steps (summary) | Expected result |
|---|---|---|---|---|
| HP-01 | New Lead | Email not in HubSpot | Submit form with unique email | New contact created; Slack alert sent |
| HP-02 | Existing Lead | Email exists in HubSpot | Submit form with existing email | Contact updated; Slack alert sent |
4. Edge & failure cases
| ID | Case | Trigger | Expected behaviour |
|---|---|---|---|
| EF-01 | API Timeout | HubSpot down | Automation retries or logs error; no Slack alert |
| EF-02 | Missing Email | Null email field | Workflow terminates; error logged in Make |
| EF-03 | Duplicate Hit | Concurrent submissions | Idempotency check prevents double creation |
5. Data / privacy / ops checks
- Test data only (or approved prod sandbox)
- No secrets in logs / screenshots
- Idempotency / duplicate event behaviour noted
- Handoff to human (if any) verified
- Other: Verify HubSpot PII encryption settings.
6. UAT execution checklist
- HP-01 Verify new record creation
- HP-02 Verify field mapping on update
- EF-01 Simulate HubSpot 503 error
- EF-02 Submit form with invalid email format
7. Go / no-go (human decision)
Ship if: All Happy Paths pass and Slack notifications contain correct deep links.
Hold if: Duplicates are created or HubSpot API keys are exposed in logs.
Retest if: Make.com scenario hits rate limits during bulk submission test.
Observed results: (fill after running — do not invent)
8. Open questions for the builder
- Is there a fallback if the Slack channel is archived or renamed?
- Are we matching duplicates by email address only or including phone numbers?
Next steps
- Review the actors and surfaces to ensure all API permissions are mapped.
- Execute the HP-01 and HP-02 tests in a HubSpot sandbox environment.
- Verify the HubSpot-to-Slack field mapping matches the client's requirements.
Connects securely to your tools. The creator never sees your data.
About this skill
Unlike generic "write some test cases" prompts that pad lists or pre-fill "all passed", this skill turns your workflow description into a client-ready UAT pack: scope, actors and surfaces, ID'd happy paths, edge and failure cases (timeouts, duplicate webhooks, auth fails, human handoff), data/privacy checks, a runnable checklist and a go/no-go block left blank for your real results. Unknown integrations stay Unknown.
Sample output (excerpt — WhatsApp order-status bot + Shopify, staging, Generated: 2026-10-16):
| ID | Flow | Preconditions | Expected result |
|---|---|---|---|
| HP-01 | Customer asks "where is my order" with order # | Test order #1042 in staging | Bot returns fulfilment status + tracking link |
| HP-02 | Customer asks for a human | Agent inbox online | Handoff message + ticket created with chat transcript |
| EF-01 | Order # not found | Typo in order # | Polite retry prompt, no internal error shown |
| EF-02 | Shopify webhook fires twice | Duplicate fulfilment event | One customer message only (idempotent) |
| EF-03 | Shopify API timeout | Simulated 10s delay | Fallback reply + retry, no blank message |
Data / privacy checks
- Test orders only; no real customer phone numbers
- Access tokens never appear in logs or screenshots
Go / no-go (human decision) Ship if: HP-01..02 and EF-01..03 pass on staging · Hold if: any duplicate customer message · Observed results: (fill after running — not invented)
Works in Claude Code, Cursor and Codex · built for chatbot, Shopify and webhook flows · no fake pass counts or coverage claims · v1.0, updated Oct 2026 · questions answered via Agensi messages · More Cygnix Labs tools: https://apify.com/cygnix_labs
Turn a chatbot, Shopify, or agent workflow into a structured UAT process template: scope restatement with unknowns; actors and surfaces; happy-path cases; edge/failure cases; data/privacy/ops checks; checkbox execution list; go/no-go criteria with blank Observed results; and open questions. Process template only — it does not run tests, provide pass/fail results, guarantee coverage, or certify compliance. Supply workflow_summary plus any optional channels_or_surfaces, actors, critical_flows, integrations, known_risks, and environment. Cygnix Labs · info@cygnixlabs.com · cygnixlabs.com · portfolio.cygnixlabs.com
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 1 month ago