A high-accuracy AI skill for verifying physical user presence through single-image passive liveness detection.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install didit-passive-liveness
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install didit-passive-liveness using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Didit Passive Liveness?
The Didit Passive Liveness skill enables AI agents to verify that a user is physically present by analyzing a single captured image. Unlike active methods, this requires no movement or interaction, providing a seamless user experience while maintaining a 99.9% accuracy rate. It is a critical component for developers building secure identity verification flows within the Openclaw Skills ecosystem.
This skill detects various presentation attacks, including screen captures, printed photos, masks, and deepfakes. By integrating this into your workflow, you can ensure that the person interacting with your application is a real human, significantly reducing the risk of fraud and unauthorized access without adding friction to the user journey.
Didit Passive Liveness Use Cases
- Preventing fraud during user registration or high-stakes transactions.
- Detecting deepfakes and presentation attacks in real-time.
- Implementing frictionless anti-spoofing checks without requiring user movement.
- Verifying face quality and luminance for better identity documentation.
- Integrating identity verification into automated agent workflows via Openclaw Skills.
How Didit Passive Liveness Works
- The agent captures or receives a user selfie in supported formats like JPEG, PNG, or WebP.
- The image is transmitted to the Didit standalone API using the DIDIT_API_KEY for authentication.
- The AI analyzes the single frame using a single-frame CNN (Passive method) to verify human presence.
- The system calculates a liveness score, assesses face quality, and performs luminance analysis.
- A final status of Approved or Declined is returned along with metadata like age, gender, and risk warnings.
Didit Passive Liveness Setup
To utilize this within your Openclaw Skills collection, you must first obtain an API key from the Didit Business Console. Once acquired, configure your environment as follows:
export DIDIT_API_KEY="your_api_key_here"
You can then initiate a liveness check using the provided utility scripts:
python scripts/check_liveness.py selfie.jpg --threshold 80
Didit Passive Liveness Data Schema & Taxonomy
The skill returns a structured JSON response detailing the liveness evaluation. Data is organized as follows:
| Field | Description |
|---|---|
| request_id | Unique session identifier for tracking |
| liveness.status | The primary result: Approved or Declined |
| liveness.score | Confidence score (0-100) where higher is more likely real |
| face_quality | Image quality metric used to filter poor captures |
| entities | Metadata including estimated age, gender, and face bounding box |
| warnings | Array of risk tags such as LIVENESS_FACE_ATTACK or DUPLICATED_FACE |
Didit Passive Liveness Advanced Features
- Configurable liveness decline thresholds to customize security sensitivity.
- Multi-platform support for JPEG, PNG, WebP, and TIFF image formats.
- Automated detection of duplicate faces across different sessions to prevent multi-accounting.
- Integrated blocklist matching to automatically decline known fraudulent identities.
- Comprehensive risk reporting including screen capture and deepfake detection within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Didit Passive Liveness Detection
A high-precision AI skill for detecting physical presence and preventing facial spoofing from a single selfie image.
Didit KYC Onboarding
Automate end-to-end identity verification including ID scanning, liveness detection, and face matching for secure user onboarding.
Didit ID Verification
Automate global identity document verification and OCR extraction within your AI agent workflows.
Didit ID Document Verification
A professional-grade identity verification skill for automating OCR, MRZ parsing, and authenticity checks across 4000+ global document types.
Didit Phone Verification
A robust identity verification skill that enables AI agents to send and validate one-time passwords across multiple messaging channels with built-in fraud prevention.
Didit Proof of Address
A specialized API skill for verifying residential addresses through automated document analysis and OCR.
Featured*
