A QA tester inside your AI
Chat with ProdPoke from Claude, Cursor, or any MCP client. It launches real browsers, finds real bugs, and remembers what you talked about.
Zero install. One config. Just start talking.
How a session looks
One conversation. Multiple scans. Full context between messages.
Scan https://prodpoke.com and give me your first impressions
The pitch lands immediately — "Paste a URL. A real browser tests your site." is strong and clear. But the stats look off: "0+ websites scanned" and "0+ pages explored" read as unfilled placeholders, which is a credibility killer. 3 network requests failed during load, and there are too many similar CTAs competing for attention. Strong concept, shaky execution.
What would you improve first if this was your product?
The "0+ websites scanned" and "0+ pages explored" statistics. Those two zeros actively undermine trust at the exact moment a visitor is deciding whether to try the product. The "100+ bugs found" surviving alongside two zero-counters makes it look like a mistake rather than an early-stage product. Fix that inconsistency and the page immediately feels more credible.
What you can ask
Just talk naturally. ProdPoke figures out what to do.
Full QA scan — bugs, SEO, performance, accessibility, security headers.
Describe what to do in plain English. A real browser executes it.
Follow up on findings. Ask why, ask for priorities, ask for fixes.
Compare against competitors. Track bugs across scans. Set preferences.
Setup in 60 seconds
Claude Desktop / Cursor / Claude Code
{
"mcpServers": {
"prodpoke": {
"type": "streamable-http",
"url": "https://prodpoke.com/v1/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Other clients (stdio bridge via mcp-remote)
{
"mcpServers": {
"prodpoke": {
"command": "npx",
"args": [
"mcp-remote",
"https://prodpoke.com/v1/mcp",
"--header",
"Authorization: Bearer ${PRODPOKE_API_KEY}"
],
"env": {
"PRODPOKE_API_KEY": "pp_your_key_here"
}
}
}
}Get your API key: Settings → API Keys — create a key and paste it into the config above. The key is shown once, so copy it immediately.
How it works
Ask anything — "scan mysite.com", "test the login flow", "what should I fix first?". Your AI calls ProdPoke automatically.
A headless Playwright browser navigates your site, clicks buttons, fills forms, and runs QA evaluation modules.
Each message in a session builds on the last. Follow up, go deeper, change focus — ProdPoke remembers everything.
Pricing
Every new account gets 500 free credits. A typical scan costs ~10 credits. Deeper workflows (experience testing, competitor comparison) cost 400-500.
Need more? Plans start at $15/month with 10,000 credits. Top-up packs available anytime.
Ready to talk to your QA tester?
Sign up, grab your API key, paste the config — you'll be chatting in under a minute.