Documentation

Introduction

Digniture.io combines AI-assisted content generation with a publishing workflow designed for short-form video. Instead of bouncing between multiple tools, the platform keeps the key steps in one place: account access, AI video generation, render tracking, saved outputs, and TikTok publishing.

The platform is currently structured around a dashboard-based workflow. Users authenticate with TikTok, maintain an account balance, create AI video jobs, review generated outputs, and optionally push completed videos into a TikTok publish flow.

Getting Started

New users typically move through the platform in this order:

Platform Overview

Digniture.io is organized as a content workflow platform. Key areas include:

Authentication

Digniture.io uses TikTok OAuth for user authentication. Users connect their TikTok account through the login flow, and the platform stores access and refresh tokens for future requests.

How authentication works

• User clicks Login or Get Started.
• The user is redirected to TikTok authorization.
• TikTok returns an authorization code.
• The platform exchanges that code for access and refresh tokens.
• The user account is created or updated in the local database.

What gets stored

• TikTok access token
• TikTok refresh token
• Token expiration timestamps
• Basic TikTok user info such as display name, username, avatar, and profile details

User Profiles

Digniture.io supports a broader user profile layer beyond TikTok data. This can include profile details, social links, privacy preferences, and optional notification settings.

Profile fields may include

• First name and last name
• Display name
• Bio
• Location
• Date of birth
• Website and social profile URLs
• Profile privacy settings

This lets the platform function as more than a simple TikTok connector. It supports a fuller creator profile and lays the groundwork for more advanced platform features later.

Account Balance

AI video generation uses a balance-based billing system. Before a new AI video job can be created, the user must have enough available funds in their account balance to cover the selected render cost.

Balance workflow

• User opens the balance page.
• User chooses a top-up amount.
• Payment is completed.
• The system credits the user balance.
• Each AI video charge is deducted from that balance.

AI Video Generation

The AI generation flow is designed to be simple from the user side while still storing enough job detail to support billing, render tracking, saved outputs, and later publishing.

Common generation fields

Generation flow

1. User fills in the AI video form.
2. The system validates the selected options.
3. The system calculates the cost based on duration.
4. The platform checks whether the user balance is high enough.
5. The user is charged before the render request is sent.
6. The platform creates a provider job and stores the job record locally.
7. The user can refresh status until the output is ready.

Pricing

AI video pricing is based on the selected duration. Longer videos cost more because they require more rendering resources and longer generation time.

The platform should display the current video price before the generate button so users can see the exact cost of the selected duration before confirming the job.

Job Status

Each AI render job moves through a status lifecycle. These states help the user understand whether a video is still in queue, currently processing, completed, or failed.

Stored job fields

• Prompt
• Style
• Duration
• Aspect ratio
• Scene count
• Provider job ID
• Status
• Result video URL
• Local saved video path
• Error message
• Raw provider response data

Generated Videos

The generated videos page is intended to act as the user’s AI video library. It should provide a simple way to view previous jobs, preview output files, refresh pending jobs, and reopen videos that are ready for posting.

Recommended display features

• Video preview player or thumbnail
• Job status badge
• Prompt summary
• Duration and format details
• Open remote result link
• Open local saved file link
• Refresh status button for unfinished jobs

This page is especially useful when users create multiple renders and want a clearer separation between generation and publishing.

TikTok Publishing Workflow

Once an AI video has completed and a usable file exists, the video can be sent into the TikTok publishing flow. This step uses the authenticated TikTok account and the TikTok publishing API.

Typical publishing flow

1. User selects a completed AI video job.
2. The platform verifies that a saved local file is available.
3. The user chooses privacy and posting options.
4. The system creates a TikTok publish initialization request.
5. A publish ID is saved for tracking.

Privacy Controls

During the TikTok posting process, the platform can expose creator-level posting controls that match available options from the connected TikTok account.

Common controls

• Privacy level
• Disable comments
• Disable duet
• Disable stitch

Some accounts may have creator-level restrictions already enabled. In those cases, the platform should respect those settings and lock the option appropriately.

Billing & Refunds

Digniture.io charges users before starting an AI video job. This helps prevent unpaid usage and keeps the rendering flow tied directly to available balance.

When users are charged

• Immediately before the render request is created

When users may be refunded

• The provider model fails to load
• The render request cannot be created
• The job setup fails before the provider job is properly stored

Replicate Setup

Digniture.io uses Replicate for AI video generation. For this to work, the platform must be configured with a valid Replicate API token and the correct provider request functions.

Required pieces

• A Replicate account
• A valid Replicate API token
• Provider request helper functions in the platform code
• Database columns for AI job tracking

Typical configuration example

Things to verify if setup fails

• Token is present and valid
• Server supports cURL
• The selected model can be requested successfully
• The AI jobs table includes all fields being inserted

Troubleshooting

AI video job does not start

• Check that the user has enough balance.
• Check that Replicate is configured correctly.
• Check for model load errors or request errors.

Video charge succeeds but no job appears

• Confirm the database table schema includes all AI job columns.
• Check for SQL insert errors in the AI jobs table.

Generated video does not open

• Verify the local file path is correct.
• Verify public file access if using a public URL.
• Check whether the file downloaded successfully after render completion.

TikTok publish flow fails

• Check that the TikTok token is valid.
• Check the privacy option selected.
• Verify the upload or public file URL is correct and reachable.

Login issues after logout

• Make sure the logout flow does not immediately redirect into TikTok login again.
• Use a logged-out state on the login page instead of auto-starting OAuth every time.

FAQ