Hacktron Automations
Review and Whitebox find vulnerabilities faster than most teams get to fix them. Automations hands each new finding that matches your rules to a Hacktron agent, which checks that the finding is real, fixes it, runs your tests and opens a pull request for your team to review.
Remediation
The agent starts by testing whether the finding can be exploited in your code. When it can’t, the code stays as it is and the run records why. When it can, the agent makes the smallest change that fixes the root cause, runs the repository’s tests and adds regression tests for the fix.
Pull requests
The pull request goes where the bug came from, on GitHub, GitLab or Azure DevOps.
Rules
An organisation admin sets an automation up once, and it runs on every match from then on.
Questions
Will it push to our code?
It pushes to a branch of its own and opens a pull request from it. It never merges.
Which findings start a run?
New findings that match an enabled automation. Each finding gets one run at most. Any member can start a run on an older finding with Remediate.
How is this different from Fix with AI in Review?
Fix with AI hands the finding to a developer’s editor. With Remediation, Hacktron’s agent makes the fix, so nobody has to pick the finding up.
What does a run cost?
Runs use automation credits, one balance for the organisation, charged when a run completes. Cost varies with the size of the run, and each environment can have a cap.