> ## Documentation Index
> Fetch the complete documentation index at: https://docs.didit.me/llms.txt
> Use this file to discover all available pages before exploring further.
# User Flow
> End-to-end Didit verification UX: ID capture, biometric liveness, smart retries, real-time feedback. Optimized to drive industry-leading completion rates.
export const AgentPromptAccordion = ({prompt, title = "AI Agent Integration Prompt"}) => {
const [copied, setCopied] = React.useState(false);
const handleCopy = e => {
e.stopPropagation();
if (!prompt) return;
navigator.clipboard.writeText(prompt.trim()).then(() => {
setCopied(true);
setTimeout(() => setCopied(false), 2000);
});
};
const agents = ["Claude Code", "Codex", "Cursor", "Devin", "Windsurf", "GitHub Copilot"];
return
{}
{title}
{}
{prompt.trim()}
{}
Paste into
{agents.map(name => {name})}
;
};
export const VideoEmbed = ({src, title = "Video", type = "iframe"}) =>
{type === "iframe" ? : }
;
This guide explains what a typical end-user experiences during an **ID Document Check + Liveness Test** session, and highlights the UX optimizations that drive high completion rates.
## Flow Overview
The user lands on a clean, branded screen that sets expectations (e.g., "You'll need your ID and your face"). This reduces drop-off by removing uncertainty.
The user selects their country and document type (Passport, ID Card, etc.).
If your workflow is configured for a **single document type**, this screen is **automatically skipped** — taking the user straight to capture.
The user photographs the front of their document.
**On-device quality check** — Before submitting, the SDK checks for glare, blur, and cutoff edges. If a problem is detected, the user is prompted to retake.
**Backend validation with guided retry** — Once submitted:
* **Pass** → proceed to liveness
* **Fail** → the user sees specific instructions (e.g., "Document expired", "MRZ unreadable — retake in stronger light") rather than a generic error
This self-serve retry loop significantly increases pass-through rates. See the full list of validation outcomes under [ID Verification Warnings](/core-technology/id-verification/warnings-id-verification).
A selfie liveness check ensures the user is physically present and prevents replay/deepfake attacks. If the user doesn't respond in time, the process **retries once automatically** instead of hard-failing.
After all checks complete, the user sees one of three results:
| Result | Description |
| ------------- | ------------------------------------------------------------------- |
| **Approved** | All checks passed. Data is extracted and available via API/webhook. |
| **Declined** | The system determined the verification cannot be approved. |
| **In Review** | Automated checks were inconclusive — requires manual review. |
By combining smart UX, progressive validation, actionable feedback, and graceful retries, the Didit verification journey keeps good users moving and keeps bad actors out — without manual intervention.
