Asterloop Product · Working product · pilot teams welcome
Asterloop Test Management
Self-hosted test management for QA teams: test cases, cycles, plans, a test player and reports, with AI test generation built in and import and export to the tools teams already use.
Screens show sample data for a fictional web shop.
Self-hosted
One Node.js server and one SQLite file, on your machine, your server or Docker.
Roles and access
Owner, Admin, Lead, Tester and Viewer roles, groups with folder access, personal API tokens.
No lock-in
Import from and export to TestRail, Xray, Zephyr Scale and CSV, with a preview before anything is saved.
Built secure
Rich text cleaned on the server, rate-limited AI and API, tested against cross-user access.
Test case library
Every test case in one place.
Folders, priorities, statuses and labels, sortable and searchable, with the columns QA teams expect. Keys such as TC-12 are never reused, and cases are archived, never deleted, so history and results stay intact.
- Clone cases as new drafts, archive and restore
- Filter by folder, label, status or priority
- Pages of 25 to 200 cases
Test case editor
Write clear, repeatable steps.
Each case has its own page with Details, Test Script, Traceability, Execution and History. Steps have rich-text step, test data and expected result fields, in a list or column layout.
- Rich text: tables, images, code and links
- Link cases to Jira issues or web pages
- Every edit is recorded with who and when
Test player
Run a cycle without losing your place.
Cases grouped by status, a result per step, a timer, the environment, comments and defects as Jira links. Keyboard shortcuts P, F, B and N record results; the case result follows the steps.
- Show only the cases assigned to me
- Start a new execution and keep the old one
- Time spent recorded per case
Test cycles and plans
See where every release stands.
Cycles show version, progress, status and the testers assigned. Plans group cycles for a release and add up their progress. Clone a cycle to rerun a regression with results reset.
- Assign cases to testers, one by one or all at once
- Planned start and end dates per cycle
- Post results from CI through the REST API
Reports
Know what passed, what failed and what's left.
Pass rate, cases not in any cycle and cases by priority, plus dashboards for versions, regression trends and test activity, each as a bar, scatter, histogram or bubble chart, or as a table.
- Spot cases that passed before and now fail
- Time spent per execution and per tester
- Filter by version, regression and period
AI test generation
From acceptance criteria to test cases in seconds.
Paste a user story, upload a screenshot or point at a Jira issue. Claude or a local model (Ollama) drafts structured test cases; skills shape the output, and you review before anything is saved.
- Text, image and Jira sources, with optional context
- Runs on a local model when data can't leave
- Ask Asterloop: a read-only assistant for your test data
AI agents (MCP)
Your coding agent can use your test cases too.
Asterloop includes an MCP server, so Claude Code, Cursor and other AI agents can read and write test cases, run cycles and record results while they work on your code.
- Uses your personal API token: the agent gets your role and folder access, nothing more
- New test cases are saved as drafts for a person to review
- Records Pass and Fail from automated runs straight into a cycle
- Works with a local model too, so test data can stay on your machine
Things to ask your agent
- "Write test cases for this story and save them as drafts in Checkout."
- "Which requirements have no tests or failing tests?"
- "The Playwright run finished: record the results in cycle CY-3."
- "Find the checkout test cases with no expected results."
See it with your own test cases.
Book a 30-minute demo, or join as a pilot team and help shape the roadmap.