Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Updated API reference URLs to docs.didit.me, added feature guide links
What This Skill Does
Compares two facial images using the Didit Face Match API and returns a similarity score from 0 to 100. Supports JPEG, PNG, WebP, and TIFF formats up to 5MB per image. Also returns estimated age, gender, and face bounding boxes alongside the match result.
Handles image rotation automatically and returns structured metadata including age and gender estimates alongside the match score, reducing manual preprocessing steps.
When to use it
- Selfie-to-ID document verification during user onboarding
- Confirming a returning user matches their registered profile photo
- Checking if two submitted photos show the same person
- Biometric access confirmation for a secure area or service
- KYC face check for financial or legal account opening
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Does this selfie match the passport photo on file?
- 1Load the selfie and passport image files
- 2POST both images as multipart/form-data to the face-match endpoint with the x-api-key header
- 3Read the face_match.score and face_match.status fields from the response
- 4Return Approved if score meets the threshold, Declined if below it
Similarity score of 80, status Approved — faces match
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.