# Continuous coverage (/docs/pentest/continuous)



A [one-time pentest](/docs/pentest/overview) tests your applications once and ends in a report. With continuous coverage, Trace keeps testing them as your code changes, so your findings stay current. You also get one official pentest report a year from an OSCP or CISSP engineer, which can stand in for your annual pentest. See [Pricing](/pricing#continuous).

How it works [#how-it-works]

You [set up your applications](/docs/pentest/applications) the same way you would for a one-time pentest. The Trace team then sets a testing schedule with you. Findings and coverage from earlier one-time pentests carry over.

Each run reads the code that changed, looks for new attack paths, and tries to exploit them against your running application. It skips anything that repeats a finding you already have. See [Verified findings](/docs/pentest/verified-findings).

What you get [#what-you-get]

* **One list of findings.** Every finding Trace has published, from every run and every one-time pentest, in one place in your dashboard.
* **A record of each run.** See what each run found and how far along it is.
* **Proof of coverage.** See which features and API endpoints Trace has tested, and which it hasn't yet.
* **Your own leads tested.** Send Trace an attack you suspect, and the next run tests it.

<Screenshot light="/docs/screenshots/continuous-findings-light.png" dark="/docs/screenshots/continuous-findings-dark.png" alt="The Continuous coverage page in the Trace dashboard, with tabs for Findings, Hypotheses, Runs, and Coverage. Findings lists 44 findings from continuous runs and one-time pentests, each row showing its ID, severity, title, category, CWE, and the run or pentest it was found in." />

Submitting hypotheses [#submitting-hypotheses]

A hypothesis is an attack you think might work against your application. Submit one, and the next run tests it. Leads can come from your own team, another security tool, or a report from an outside researcher. Each one goes through the same checks as Trace's own leads, so you have one place to triage them all.

Submit leads with the [Trace CLI](/docs/cli/overview#tracecli-pentest-hypothesis-submitlistwithdraw):

```bash
tracecli pentest hypothesis submit \
  --title "IDOR on /exports/:id" \
  --description "Any signed-in user can fetch another tenant's export by ID."
```

Fixing and retesting findings [#fixing-and-retesting-findings]

Every finding comes with a full remediation recommendation, with references to the code that needs to change. Mark a finding **Remediated** once your fix is deployed, and Trace retests it. There's no limit on retests. You can also accept or dispute a finding. See [How an engagement works](/docs/pentest/engagement#fixing-findings).
