What is Claudit and what does it do?

Claudit is a tool that integrates with AI coding agents like Claude Code and Codex CLI, enabling them to search Solodit's extensive database of smart contract security findings for vulnerability insights.

Which AI coding agents does Claudit support?

Claudit is designed to integrate seamlessly as an MCP server for both Claude Code and Codex CLI, providing these agents with powerful search capabilities across Solodit's database.

What kind of search filters are available in Claudit?

Claudit offers extensive filtering options including severity, audit firms, vulnerability tags, programming language, protocol, quality score, rarity score, user, number of finders, and reported dates.

Is a Solodit API key required to use Claudit?

Yes, a valid Solodit API key is required to access the database of security findings. It needs to be configured during the installation process or manually.

Can I retrieve full details for a specific security finding?

Yes, Claudit allows you to retrieve complete details for any specific security finding by providing its numeric ID, Solodit URL, or slug.

Claudit: Smart Contract Security Findings for AI Agents

Claudit functions as an MCP server, integrating directly with AI coding agents like Claude Code and Codex CLI to unlock access to Solodit's vast repository of over 20,000 smart contract audit findings. It empowers these agents to make more informed decisions by allowing them to search, filter, and retrieve detailed information on past vulnerabilities. This tool is designed to enhance the security awareness and analysis capabilities of AI-driven development workflows, offering rich context on various types of findings, audit firms, protocols, and more.

主要功能

01Search Solodit findings with extensive filters for severity, firms, tags, language, and protocol.

02Retrieve full details for specific security findings using numeric ID, Solodit URL, or slug.

03Access a list of all valid filter options, including firms, tags, categories, and languages, with finding counts.

04Seamless integration as an MCP server for Claude Code and Codex CLI.

05Supports advanced filtering based on quality score, rarity score, user, number of finders, and reported dates.

06110 GitHub stars

使用案例

01Provide AI coding agents with a comprehensive knowledge base of past audit findings to improve code generation and vulnerability detection.

02Empower AI agents to identify specific smart contract vulnerabilities (e.g., reentrancy, oracle manipulation) across various protocols.

03Enable developers and AI to research detailed security findings by audit firm, auditor, or historical context.