AI-Powered SecurityLiveFrom Builder ($69/mo)

AI Remediation

Generate an AI-written fix for a finding from any engine. Each fix explains the root cause, gives the code, config or commands to apply, states how confident it is and includes a command to verify the result. You review it before anything changes.

Detection Coverage

Fix types AI Remediation generates

Code Patches

Code fixes for application findings, such as parameterised queries, input validation and output encoding.

Server Config

Security header and TLS configuration for DAST and health check findings.

Dependency Updates

The package and version to upgrade to for SCA findings.

Secret Rotation

Steps to revoke and rotate a leaked credential.

IaC and Cloud Config

Corrected infrastructure-as-code blocks and cloud configuration changes for IaC and CSPM findings.

Image Hardening

Changes to make to a container image for container findings.

AutoFix Pull Requests

For a connected GitHub repository: a dependency bump for SCA findings, or an AI-generated diff for SAST findings, opened as a pull request.

AI False Positive Review

An AI review of SAST findings that lowers the confidence of likely false positives.

How it works

From finding to verified fix

  1. 1

    Pick a finding

    Open a finding from any engine and ask for an AI fix.

  2. 2

    AI generates the fix

    The fix is written for the finding type and your tech stack. Secrets in the evidence are redacted before the request leaves Spoorder.

  3. 3

    Review it

    Read the explanation, the fix and its confidence level: high, medium or low. Nothing is applied for you.

  4. 4

    Apply and verify

    Apply the fix, run the verification command, then mark the finding as remediated. A re-scan confirms whether it is gone.

ai-fix-diff.patchdiff
--- a/src/api/users.ts
+++ b/src/api/users.ts
@@ -14,3 +14,5 @@
- const q = `SELECT * FROM users WHERE id = ${id}`;
- await db.query(q);
+ const q = 'SELECT * FROM users WHERE id = $1';
+ await db.query(q, [id]);
FindingExample AI fix for SQL injection (CWE-89): a parameterised query replaces string interpolation.

Spoorder AI Remediation vs researching each fix by hand

CriterionSpoorderresearching each fix by hand
Fix coverageFindings from any engineWhatever the engineer already knows
Fix formatCode, config or commands with an explanationA link to documentation
ConfidenceStated as high, medium or lowNot stated
VerificationA verification command, then a re-scanManual
AutoFix PRsGitHub pull requests for SCA and SAST findingsManual implementation
Does AI Remediation work on all engines?

Yes. You can generate a fix for a finding from any engine. The type of fix depends on the finding: a code patch, a server or cloud config change, a dependency update, secret rotation steps, an IaC patch or image hardening.

How accurate are the AI fixes?

It varies with how much evidence the finding carries, which is why every fix states its confidence as high, medium or low and explains why. Review every fix before you apply it, and re-scan afterwards.

Is my data sent to an AI provider?

The finding details needed to write the fix are sent to the AI provider. Tokens, keys, passwords and similar secrets in the evidence are redacted first.

How many AI fixes do I get?

Builder includes 10 AI fixes per month, Pro 100 and Scale 500. Enterprise is custom. The Free plan has none.

Do AutoFix pull requests need anything set up?

Yes: a connected GitHub repository that Spoorder has permission to open pull requests on. AutoFix covers SCA findings (a dependency bump) and SAST findings (an AI-generated diff).

Start scanning

Start with a free health check. AI Remediation is included from Builder ($69/mo).