Settings reference

All organization settings live under the Settings section in the left sidebar. Changes apply to your current organization only.

LLM configuration

/settings/llm

API: GET / PUT /api/settings/llm

Configure the model used for AI-assisted SAST, LLM secret detection, zero-day analysis, Suggest fix, and Open fix PR file rewriting.

Fields

FieldDescription
ProviderPreset: OpenAI, OpenRouter, Azure, Ollama, or custom. Sets default base URL and model.
Base URLOpenAI-compatible API root (e.g. https://api.openai.com/v1, https://openrouter.ai/api/v1, http://localhost:11434 for Ollama).
ModelModel ID sent to the API (e.g. gpt-4o-mini, google/gemini-2.5-flash, qwen2.5-coder:7b).
API keyStored encrypted per organization. Leave blank on save to keep existing key. Env fallback: LLM_API_KEY, OPENAI_API_KEY, OPENROUTER_API_KEY.
Enable LLM SASTWhen on, full scans use LLM code analysis instead of pattern-only SAST; enables zero-day scanner on full scans.
Enable LLM secretsWhen on, secrets scanner uses LLM in addition to patterns.
OSV API URLSoftware composition / vulnerability database endpoint (default OSV.dev).
Vuln DB modeonline (query OSV) or offline mode when air-gapped.
Ollama tip: Run Ollama on the host; set base URL to your Ollama host. Use README model table for memory vs quality tradeoffs.

Security policies

/settings/policies

API: GET/POST /api/settings/policies, PATCH/DELETE /api/settings/policies/[policyId]

Organization-wide rules expressed in natural language. LLM SAST uses enabled policies as extra context when analyzing code.

Creating a policy

Use example templates on the page to seed common policies.

Build gates

/settings/build-gates

API: PUT /api/settings/build-gates with projectId

Per-project thresholds evaluated when a scan completes. Result stored on the scan as gateResult: PASS, FAIL, or WARN.

Fields

FieldDescription
Max criticalFail if critical count exceeds this (default 0 = no critical allowed).
Max high / medium / lowUpper limits per severity. Use -1 for unlimited (UI empty = -1).
Fail on newWhen enabled, fail if any finding is new compared to the previous completed scan on the same project.

Select a project from the dropdown, adjust thresholds, click Save. View gate status on the project page and scan list.

Team

/settings/team

API: GET/POST /api/users, PATCH/DELETE /api/users/[memberId]

Manage organization membership and roles.

Roles

RoleTypical use
ADMINFull org settings, team management
SECURITYRun scans, triage findings, configure policies
DEVELOPERView and update finding status, open fix PRs
VIEWERRead-only access

Invite flow

  1. Enter email and role, send invite.
  2. New users receive credentials by email when SMTP is configured.
  3. Existing users are added to the org immediately.

Integrations

/settings/integrations

Connect external systems and automate scans.

GitHub (OAuth)

No personal access token UI — OAuth only for fix PRs.

GitLab webhook

GitHub webhook

POST /api/webhooks/github — set GITHUB_WEBHOOK_SECRET in env and configure the secret in your GitHub repo/org webhook.

CI / API key

The integrations page shows a sample curl to POST /api/scans with an API key header. Create API keys via your deployment’s admin/seed process or future UI.

Project settings (not under Settings menu)

/projects/[projectId]/settings

Edit project name and description only.

Scan schedule is not in this UI — use PUT /api/projects/{projectId}/schedule (see README).