Free trial
The first 10 completed runs are free and shared across the workspace. Runs that cannot finish because of an agent or platform failure do not use the free allowance.
Learn morePlain-English end to end tests that run in parallel and find gaps
Own Wren by Seldon? Verify ownership to get a dofollow link
Wren by Seldon is a developer tool. Wren by Seldon runs plain-English end to end tests in parallel and uses behavioral models to explore paths a suite misses, with recorded steps to reproduce each failure. It costs $3 per completed run after 10 free runs.
Wren by Seldon is a behavioral software testing tool aimed at engineers and QA teams who maintain end to end browser tests. Instead of scripting every click and assertion in a framework such as Playwright, a tester describes a task in a sentence or two, says where it starts and what a successful result looks like, and Wren carries it out in a browser. The workflow has three stages. First, the test is written in plain English, much as it would be explained to a teammate. Second, runs execute in parallel, with each agent choosing its next action based on what it sees on the page. If a button gives no feedback, an agent may retry it or look elsewhere. Third, the results are reviewed as recorded runs, step by step, with evidence attached to every failure and the steps needed to reproduce it. The site's own example shows a shopper clicking "Add to bag", seeing no confirmation and clicking again, which leaves two items in the cart. That outcome suggests a new test case about whether the shopper can tell an item was added. This exploratory behavior is the main difference from conventional scripted suites. Seldon's behavioral models guide what each agent tries next, so one defined workflow can surface paths the team never thought to write down. The listing's own comparison with Playwright presents the plain-English definition as far shorter than the equivalent script, which reduces the maintenance burden of keeping selectors and steps current. Pricing is usage based. The first 10 completed runs are free and are shared across the workspace, and later completed runs cost $3 each. Model usage, browser execution, findings and recorded evidence are included, there are no seat fees, and runs that fail because of an agent or platform problem are not charged. Wren fits teams that want to run it alongside an existing suite on a single flow and compare what it explores. It complements rather than replaces deterministic regression scripts, and it suits teams asking what a release broke more than those needing exact repeatable assertions.
The first 10 completed runs are free and shared across the workspace. Runs that cannot finish because of an agent or platform failure do not use the free allowance.
Learn moreThe first 10 completed runs are free, shared across the workspace. After that it costs $3 USD per completed test run. Model usage, browser execution, findings and recordings are included, with no seat fees or token charges. Runs that fail due to the agent or platform are not charged.
Learn morePlain-English test definitions, parallel agent runs, behavioral models that guide exploration, recorded step-by-step runs, evidence for each failure, and steps to reproduce issues.