mcpbeat

Manual Testing

datadog/manual-testing

Run a manual test of the current change end-to-end and output reproducible test instructions for the PR "Test instructions" section.

569 tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
396
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository 
npx skills add https://github.com/DataDog/browser-sdk --skill manual-testing

The instruction itself

6 sections, as written by the author

Step 1: Understand the change

If you don't have the PR changes in memory, review all commits in the PR to identify what SDK behavior changed and what events/fields need to be verified.

Step 2: Start the dev server

The dev server serves the sandbox/ directory and proxies intake requests locally. Use yarn dev-server --help to list all available commands.

yarn dev-server start

Step 3: Create a temporary sandbox page

Create sandbox/test-<topic>.html. See sandbox/index.html for a minimal example. Include only the elements needed to exercise the change. Always use proxy: '/proxy'.

cat > sandbox/test-<topic>.html << 'EOF'
<!doctype html>
<html lang="en">
  <head>
    <meta charset="UTF-8" />
    <title>Test <topic></title>
    <script src="/datadog-rum.js"></script>
    <script>
      DD_RUM.init({ clientToken: 'xxx', applicationId: 'xxx', proxy: '/proxy', trackUserInteractions: true })
    </script>
  </head>
  <body>
    <!-- elements needed to exercise the change -->
  </body>
</html>
EOF

Step 4: Run the test flow

Get the dev server URL from yarn dev-server status. Clear any previous intake data, open the page, interact with it using CSS selectors, flush events by opening a new tab or reloading, then inspect the intake:

yarn dev-server intake clear
agent-browser open <dev-server-url>/test-<topic>.html
agent-browser click '#...'
agent-browser tab new
yarn dev-server intake <selector> | jq '<field>'

Use yarn dev-server intake --help to find the right selector.

To evaluate JavaScript on the page (e.g. to inspect window state set by beforeSend), switch focus back to tab 0 first, then use agent-browser eval:

agent-browser tab 0
agent-browser eval 'window.someValue'

Step 5: Verify the output matches expectations, then present the test instructions

Output a self-contained bash snippet with the exact commands run and the expected output. This goes directly into the PR "Test instructions" section.

Step 6: Clean up

yarn dev-server stop
rm sandbox/test-<topic>.html

How to use it

Copy the folder

Take datadog/manual-testing from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.