Skip to main content

Workflow overview

Hyrax's product surface is a small set of jobs you can run against any repo. Each one has a specific job to do; most teams use all of them in a regular cadence.

WorkflowWhat it doesCostPage
AuditFind what's worth fixing (findings + a synthesis pass of suggestions)$1–35Two depths — fast (scanner) or standard (scanner + reasoning agents)
ImproveFind what could be better (suggestions only)~$7–11Same engine as audit, deeper standalone pass focused on opportunities
FixOne finding → one PR~$1–1014-step fix gauntlet, isolated worktree, reviewer agent, tests
TaskPlain-English directive → PR~$1–10Same gauntlet as fix, but driven by an instruction instead of a finding
PR reviewAI on every pull request~$1–2.50Scanner mode (inline comments) or LLM mode (rolling checklist)
DiscoveryKnowledge bundle into the repo~$5–10HYRAX.md + .hyrax/discovery/ — read by Hyrax and your IDE AI
LearnHyrax getting better at your code~$0.50–1.00Background — proposes virtual patches you review

Cadence most teams use

WhenWhat
Once, on adding a repoDiscovery (auto-runs)
Weekly or after each releaseAudit (standard)
Whenever you spot a finding worth acting onFix
When you have a goal that doesn't map to a findingTask
Quarterly or per planning cycleImprove
On every PR (automatic)PR review
Quarterly or after a refactorRe-run Discovery
Continuously (automatic)Learn

How they fit together

Findings are the unit of work — produced by audit (kind=finding) and improve (kind=suggestion), consumed by fix, tracked through tickets, closed by re-run. Discovery is the context layer — its output makes every other workflow on the repo sharper. Related repos extend that context across repo boundaries. Learn is the meta layer — it watches the others run and proposes improvements to Hyrax itself.