v25.11.01

https://hypofuzz.com/

• Drop support for Python 3.9, which reached end of life in October 2025.

v25.09.02

https://hypofuzz.com/

• Fix error while running stateful tests
• Support middle click to open links in new tab in more places

v25.09.01

https://hypofuzz.com/

• Improve coverage graph usability
• Smooth coverage curve under the "together" coverage graph view, avoiding jitter from concurrent workers
• Fix computation of covering patches

v25.08.02

https://hypofuzz.com/

• frontend UI improvements, particularly on the coverage graph
• compatibility with newer Hypothesis versions

v25.08.01

https://hypofuzz.com/

• Compute and display coverage stability of the rolling observations on the dashboard.
  • The coverage stability of a test is the percentage of inputs to that test which result in identical coverage when replayed.
  • Low stability reduces the effectiveness of mutation in coverage-guided fuzzing. HypoFuzz communicates this metric to you so you can make an informed decision about when to invest in improving stability.
• Add a new status Skipped dynamically on the dashboard, for when a test raises a skip like pytest.skip dynamically
• Improve Workers page UI

v25.07.02

https://hypofuzz.com/

Add support for unittest and class-based pytest tests. See https://hypofuzz.com/docs/compatibility.html for details.

v25.07.01

https://hypofuzz.com/

Fix a bug where the @reproduce_failure decorator would not be displayed on the dashboard for failures.

v25.06.04

https://hypofuzz.com/

Add support for pytest fixtures. This includes support for all fixture scopes, including scope="session". Fixtures are set up when starting to fuzz a test, and torn down when switching to fuzz another test.

v25.06.03

https://hypofuzz.com/

• Add new display options for the test coverage graph: Together, Separate, and Latest.
  • Together shows a linearized view of all workers for this test, even when they were concurrent in practice.
  • Separate shows each worker run for this test individually.
  • Latest shows only the latest worker for this test.
• Improve coverage calculation, to make reporting unstable coverage (due to e.g. caching) less likely. In the future, we will report coverage stability on the dashboard as a per-test percentage.