README.md

<p align="center">
  <a href="https://solidityscan.com/" target="_blank">
    <img src="https://cdn-uploads.huggingface.co/production/uploads/685934bed73d6881647be295/CL6wjrU6gKYv3_Itc4iAO.png" alt="SolidityScan CredShields Banner" width="75%">
  </a>
</p>

### CredShields on Hugging Face
[CredShields](https://credshields.com/) blends AI automation with senior-expert auditing to secure smart contracts, protocols, and enterprise applications. We focus on predictive security, developer-first workflows, and standardized risk communication inspired by OWASP. Our flagship is [SolidityScan](https://solidityscan.com/), built to catch logic flaws before mainnet while supporting human-in-the-loop review.

### Mission
- Predict and prevent exploits before deployment.
- Standardize risk across ecosystems with OWASP-style clarity.
- Accelerate secure shipping with developer-first tooling and automation.

### What We Publish Here
- Models: Solidity-aware LLMs for code understanding, and reasoning.
- Spaces: Lightweight demos for instant scans, diff-based fixes, and rule explainability.

### Flagship: SolidityScan
SolidityScan powers fast, predictive scanning for developers. It augments manual reviews with AI-driven findings, risk normalization, and actionable remediation guidance—reducing audit cycle time without compromising depth.

### Responsible Use
- These models assist auditors and developers; they do not replace manual review.
- Findings require human validation and protocol-level context before action.

### Governance and Contributions
- Use Hugging Face discussions for issues, feature requests, and disclosures.
- We follow semantic versioning and provide changelogs and reproducible evals.
- Contributions are welcome.

### Licensing
- Models: permissive licenses (e.g., Apache-2.0) unless otherwise specified in each repo.
- Datasets: open licenses suitable for research and commercial audit workflows.
- Please check each model/dataset card for exact terms.