These badges represent VIBES standard compliance levels — from simple AI usage percentages to cryptographically verified attestations. Add them to your README to communicate transparency about AI involvement in your codebase.
Badge Specification
Use shields.io badges to communicate AI usage in your projects. Each badge tier represents a different level of AI involvement.
Light AI assistance
Moderate AI assistance
Mixed development
Primarily AI-driven
Mostly AI-generated
Fully AI-generated
Additional badge styles:
General AI-Generated (blue)
AI-Assisted (flat)
General AI-Generated (dark)
Attestation Level Badges
Every AI usage claim has a confidence level. Use these badges to signal how your AI percentage was verified. Attestation badges work alongside the Low, Medium, and High VIBES assurance levels to provide full transparency. Learn more about the cryptographic attestation system.
Attestation badges are validated through the VERIFY extension. See the VERIFY specification for details on cryptographic attestation, trust tiers, and the public registry.
Developer's own estimate
Confirmed by human review
Verified by automated tooling
Combine a percentage badge with an attestation badge for full transparency:
Add Badges to Your README
Copy and paste the markdown snippet below into your README.md file. Replace the percentage with your project's AI usage level.
Percentage badge:
[](https://itsavibe.ai)
General badge:
[](https://itsavibe.ai)
AI-Assisted badge:
[](https://itsavibe.ai)
Self-Attested badge:
[](https://itsavibe.ai)
Manually Validated badge:
[](https://itsavibe.ai)
Machine Validated badge:
[](https://itsavibe.ai)
Self-Attestation Logos
Help build a more transparent AI coding ecosystem by using our itsavibe.ai logo in your README.md files. Let others know when your project is AI-generated or AI-assisted.
Preview of our logos:
400x400 version
Full size version
Add this to your README.md:
This project contains AI-generated or AI-assisted code. Learn more at itsavibe.ai
Attestation Levels
Every project in the registry declares an attestation level indicating how the AI usage percentage was determined. These levels combine with VIBES assurance levels (Low, Medium, High) to give a complete picture of AI provenance. See the attestation system for how cryptographic signing works.
Self Attestation
The project maintainer estimates the AI usage percentage based on their own assessment. This is the default level and the easiest starting point. Consider factors like lines of code generated by AI, architectural decisions, and post-generation editing.
Manually Validated
A human reviewer analyzes tool-generated audit output — provenance reports, code analysis results, and git history data — to confirm the AI usage claim. This adds credibility through informed third-party validation based on actual tool output rather than opinion alone.
Machine Validated
Automated tooling has analyzed the codebase to determine the AI-generated percentage. This is the highest confidence level, using code analysis tools, git history analysis, and AI detection models to verify claims programmatically.
How to Estimate Your AI Percentage
Consider these factors when estimating your AI percentage:
- Lines of code generated by AI vs. written manually
- Architectural decisions made by humans vs. AI suggestions
- Amount of post-generation editing
- AI's role in debugging and optimization
Remember, transparency builds trust in the AI coding ecosystem!
Related Standards
Badges communicate VIBES compliance at a glance. For cryptographic proof behind the badge, see VERIFY. For automated risk scoring that can drive badge assignment, see PRISM. For agent governance policies that inform attestation strategy, see EVOLVE.
