On demand
Prove it on one project.
$90
for 10,000 credits
- About 20 runs
- $9 per additional 1,000 credits
- Reproducible bugs sent to your coding agent
RunStory pricing
Cheaper than your next QA hire. Catches what code review misses.
Stop wasting time and tokens fixing AI code review nits. RunStory uses your software like a person would and sends real bugs back to your agent.
Plans
Team
Most teamsTest every change before it ships.
$650
a month, with 100,000 credits
- About 200 runs a month
- $65 per additional 10,000 credits
- Help teammates jump straight into your PR
Enterprise
Continuous testing.
Custom
for unlimited testing
- Unlimited runs
- 24x7 testing
- Scale quality across your org
Every plan starts with a 15-minute setup with a founder. Free evaluation credits are included.
What’s different? Review loops never end. RunStory tests what matters.
Every review round finds more to fix, and every fix costs time and tokens. RunStory brings a user’s view into your automated testing.
Illustrative example
The review agent and the coding agent go back and forth over small style comments. Every round finds more, and no user-facing bugs are found.
RunStory writes a test plan from the intent of the change, runs the software, finds a P1 bug where the cart empties, sends it to the coding agent, and verifies the fix.
Live in 15 minutes.
Book 15 minutes with a founder.
Pick a time that works for you. Your first 10,000 credits are on us.
Easy to set up.
A simple CLI your agent can call. You do not write test scripts.
Run your first test together.
You see how RunStory works on your software, in your own environment.
How credits work.
- Credits pay for the AI that tests your software.
- A small change takes less testing than a full user journey, so it uses fewer credits.
- Every run uses credits.
- That includes the re-run that checks your agent’s fix.
- Set limits.
- Set a daily or monthly credit limit. You stay in control of what you spend.
- Every bug comes with proof.
- Each finding goes to your coding agent with a full report and a deterministic reproduction script.
Compare it to your next QA hire.
Your next hire costs about $10,000 a month.
RunStory Team costs $650 a month for about 200 runs. That is 94% less. Need more runs? Add 10,000 credits for $65.
Questions.
What is a run?
A run is one RunStory test session. RunStory reads what you meant, uses your software, and reports what it finds. A run uses about 500 credits.
Do you charge per seat?
No. Your whole team can use RunStory. You pay for credits, not seats.
How is this different from AI code review?
Code review reads the diff. RunStory runs the software. In our study, it found 2 more P1/P2 bugs per 10,000 lines than GitHub Copilot, CodeRabbit, and Greptile, with under 10% false positives.
How is this different from test automation tools?
Scripted test tools run the tests you write and keep up to date. RunStory writes its own exploratory test plan from your prompts, specs, and requirements. Then it uses your software the way a person would.
Where does RunStory run my software?
In its own sandbox, separate from your systems. You keep working while it tests.
Why is there a setup call?
We tune RunStory’s sandbox to match your app, so it tests your software the way your customers use it. A founder does this with you in 15 minutes. Then we keep working with you until RunStory is finding real bugs.
Will this replace my QA team?
It gives them leverage. RunStory runs exploratory tests on every change, so your QA people can spend their time on judgment calls.
See it on your own software.
Book a setup call with a founder. You run your first test together and leave with free evaluation credits.