> For the complete documentation index, see [llms.txt](https://verogram.gitbook.io/whitepaper/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://verogram.gitbook.io/whitepaper/chapter-7-verification-system.md).

# Chapter 7: Verification System

### 7.1 Verification Philosophy

Verogram’s verification is not based on blind trust or simple screenshots. The system uses a **confidence-based** approach where each task completion is assigned a confidence score rather than a binary pass/fail.

### 7.2 Verification Methods

* **API-Based Verification:** Direct data retrieval from social platform APIs (where available)
* **Data Service Verification:** Using third-party data services for platforms with limited API access
* **Challenge-Based Verification:** Requiring users to perform specific in-platform actions to prove account ownership
* **Time-Based Rechecks:** Re-verifying task completion after a defined period to detect reversals (e.g., unfollowing after receiving reward)

### 7.3 Confidence Scoring

Instead of approving or rejecting a task in absolute terms, the system calculates a confidence score:

* High confidence → Full reward
* Medium confidence → Partial reward or pending status
* Low confidence → Rejection or manual review queue

This approach reduces false positives while maintaining fairness.

### 7.4 Reversal Detection

Some users may complete a task (e.g., follow) and then reverse it (e.g., unfollow). Verogram addresses this through:

* Scheduled rechecks at defined intervals
* Reward clawback mechanisms for detected reversals
* Risk score increases for users with reversal patterns