<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Pranshu on Medium]]></title>
        <description><![CDATA[Stories by Pranshu on Medium]]></description>
        <link>https://medium.com/@pranshu952876?source=rss-34b0fb784046------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*iw3hN1GZYY0YQ-Hb</url>
            <title>Stories by Pranshu on Medium</title>
            <link>https://medium.com/@pranshu952876?source=rss-34b0fb784046------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 08:46:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@pranshu952876/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[My Final Internship Project at Ether Authority: Smart Contract FixGPT]]></title>
            <link>https://medium.com/@pranshu952876/my-final-internship-project-at-ether-authority-smart-contract-fixgpt-dcd8da88d00f?source=rss-34b0fb784046------2</link>
            <guid isPermaLink="false">https://medium.com/p/dcd8da88d00f</guid>
            <category><![CDATA[ethereum-blockchain]]></category>
            <category><![CDATA[foundry]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[slither]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Pranshu]]></dc:creator>
            <pubDate>Wed, 14 Oct 2026 06:20:20 GMT</pubDate>
            <atom:updated>2026-10-14T06:20:20.881Z</atom:updated>
            <content:encoded><![CDATA[<p>I’m happy to share that I have completed my internship at <strong>Ether Authority</strong>, where I developed <strong>Smart Contract FixGPT</strong> as my final internship project.</p><p>During this project, I worked on building an AI-assisted platform that helps developers identify potential Solidity smart contract vulnerabilities, generate suggested fixes, and verify whether those fixes address targeted security issues.</p><p>Smart contract security is an important part of Web3 development because vulnerabilities can lead to financial losses, unexpected contract behavior, and other security risks. My goal was to build a practical workflow combining static analysis, AI-assisted remediation, compiler checks, and automated security testing.</p><h3>What Is Smart Contract FixGPT?</h3><p>Smart Contract FixGPT is a web application designed to help developers analyze Solidity contracts and investigate potential security vulnerabilities.</p><p>Rather than stopping at vulnerability detection, the application provides a workflow for proposing fixes and checking the resulting code.</p><p>The workflow includes:</p><ol><li>Analyzing Solidity source code using Slither.</li><li>Presenting reported security findings in a more readable format.</li><li>Using Gemini AI to generate suggested remediation code.</li><li>Preparing candidate code and a diff showing the proposed changes.</li><li>Compiling the candidate code using a compatible Solidity compiler.</li><li>Running Slither analysis again to check whether the targeted finding remains.</li><li>Running Foundry-based security tests for supported verification scenarios.</li><li>Generating a report containing findings, suggested resolutions, a project score, and PASS/FAIL status.</li></ol><p>This approach provides more evidence than relying on AI-generated code alone. However, resolving one targeted finding does not prove that the entire contract is free of vulnerabilities.</p><h3>Key Features</h3><h3>1. Solidity Vulnerability Detection</h3><p>Smart Contract FixGPT uses <strong>Slither</strong> for static analysis of Solidity source code.</p><p>The application organizes security findings into readable titles, descriptions, and recommended resolutions, making the results easier to investigate.</p><p>Examples of detector categories handled or surfaced by the project include:</p><ul><li><strong>Reentrancy (</strong><strong>reentrancy-eth):</strong> Identifies patterns associated with potential reentrancy risks involving Ether transfers.</li><li><strong>tx.origin misuse (</strong><strong>tx-origin):</strong> Flags potentially unsafe authorization patterns that rely on tx.origin.</li><li><strong>Low-level calls (</strong><strong>low-level-calls):</strong> Highlights low-level call operations that may require additional security review.</li><li><strong>Unchecked ERC-20 transfers (</strong><strong>unchecked-transfer):</strong> Identifies token-transfer patterns where return values may not be checked appropriately.</li></ul><p>These detector categories help developers investigate potential issues. A reported finding does not automatically mean that a contract is exploitable, and the absence of a finding does not guarantee security.</p><h3>2. AI-Assisted Remediation with Gemini</h3><p>After identifying a finding, the application uses <strong>Gemini AI</strong> to propose a potential fix.</p><p>The remediation workflow generates candidate Solidity code and a diff so developers can inspect the proposed changes rather than blindly replacing their original contract.</p><p>AI-generated code can contain mistakes or introduce new problems. Therefore, suggested fixes should be reviewed and tested before being used in a real project.</p><h3>3. Dynamic Solidity Compiler Selection</h3><p>Different Solidity projects may require different compiler versions through their pragma statements. Using an incompatible compiler can prevent a contract from compiling successfully.</p><p>To address this challenge, I implemented a dynamic compiler manager in analyzer/compiler_manager.py.</p><p>The compiler manager parses supported Solidity pragma formats, including exact versions, caret constraints, and version ranges, to select a compatible compiler version. The workflow can also install a required compiler version when it is missing from the environment.</p><p>This makes the analysis and verification workflow more adaptable to Solidity projects using different compiler requirements.</p><h3>4. Compilation and Security Re-Analysis</h3><p>After generating a candidate fix, the application attempts to compile the modified code and run static analysis again.</p><p>The objective is to check whether the targeted vulnerability is still reported and identify other findings that remain.</p><p>For example, one tested verification flow removed a targeted reentrancy-eth finding while still reporting low-level-calls.</p><p>This demonstrates an important principle: addressing one vulnerability does not automatically resolve every security issue in a smart contract.</p><h3>5. Foundry-Based Security Verification</h3><p>Smart Contract FixGPT uses Foundry tests to check contract behavior in supported scenarios.</p><p>The verification flows I tested included:</p><ul><li>A reentrancy-related security test for a sample contract.</li><li>A tx.origin authorization test designed to check whether the targeted unsafe pattern was blocked.</li><li>Solidity compilation checks.</li><li>Post-remediation Slither analysis.</li></ul><p>These checks provide additional evidence about the proposed changes. They do not replace comprehensive contract-specific testing, manual review, or a professional security audit.</p><h3>6. Reports, Resolutions, and Project Status</h3><p>The application generates reports in JSON and Markdown formats.</p><p>The report workflow includes:</p><ul><li>Detected findings and readable vulnerability descriptions.</li><li>Suggested resolutions.</li><li>Verification information where available.</li><li>An application-specific project score.</li><li>PASS/FAIL status.</li></ul><p>The score is an indicator defined by the application, not an industry-standard smart contract audit rating. PASS/FAIL results must also be interpreted in the context of the checks that were actually executed.</p><h3>Technology Stack</h3><p>The project combines blockchain security tools, a Python backend, AI integration, and a web frontend.</p><p>TechnologyPurposeSoliditySmart contract source code being analyzedSlitherStatic analysis and vulnerability detectionFoundryCompilation, execution, and security testingPythonBackend analysis and verification logicFastAPIBackend APIReactFrontend user interfaceViteFrontend development and build toolingGemini AIAI-assisted vulnerability remediationsolc / solc-selectSolidity compilation and compiler-version managementGit and GitHubVersion control and source code managementVercelFrontend deploymentRenderBackend deployment</p><h3>Solidity Compiler Versions Tested</h3><p>I tested the dynamic compiler workflow with the following Solidity compiler versions:</p><ul><li>0.8.24</li><li>0.8.28</li><li>0.8.33</li><li>0.8.37</li></ul><p>The compiler-manager tests confirmed exact-version selection for 0.8.28 and compatible selection behavior for the tested caret and version-range constraints.</p><p>These results apply to the compiler-management scenarios I tested. They do not establish compatibility with every Solidity contract, dependency, pragma format, or compiler version.</p><h3>Testing and Validation</h3><p>During development, I ran focused tests to validate important parts of the workflow.</p><ul><li><strong>Compiler manager tests:</strong> Passed for the tested compiler-selection scenarios.</li><li><strong>tx.origin verification tests:</strong> Passed for the targeted test case.</li><li><strong>Reentrancy verification:</strong> The Foundry-based test passed for the sample contract.</li><li><strong>Security verification flow:</strong> Compilation succeeded, and the workflow checked whether the targeted reentrancy finding remained.</li><li><strong>Post-remediation analysis:</strong> The application reported findings that remained after a targeted finding was removed.</li></ul><p>These tests helped validate individual components of the platform. They should not be interpreted as proof that the application detects or fixes every possible smart contract vulnerability.</p><h3>Deployment and Source Code</h3><p>Smart Contract FixGPT has a deployed web application and a public source code repository.</p><p>The live application is hosted using Vercel for the frontend and Render for the backend. The source code is maintained on GitHub.</p><p>The project can be explored through my professional profile and project portfolio.</p><h3>What I Learned</h3><p>Building Smart Contract FixGPT gave me practical experience connecting AI-assisted development with established smart contract security tools.</p><p>Some of the most valuable areas of learning included:</p><ul><li>Solidity static analysis and vulnerability investigation.</li><li>Dynamic compiler selection and Solidity version compatibility.</li><li>FastAPI backend development.</li><li>React frontend integration.</li><li>Foundry-based testing and verification.</li><li>AI-assisted code remediation.</li><li>Application deployment and debugging.</li><li>Interpreting security findings and communicating verification results.</li></ul><p>One of my biggest takeaways was that generating a fix is only one step in the security process. A useful remediation workflow should also compile the proposed code, rerun relevant checks, test expected behavior, and report any remaining issues clearly.</p><h3>Final Thoughts</h3><p>Completing my internship at <strong>Ether Authority</strong> and developing Smart Contract FixGPT as my final project has been a valuable learning experience.</p><p>This project allowed me to apply my blockchain development knowledge to a practical problem while exploring how AI can support Solidity vulnerability analysis and remediation.</p><p>I’m grateful to Ether Authority for the opportunity and for the experience I gained throughout this project.</p><p>I’m looking forward to continuing my journey in Blockchain and Web3 development, improving my smart contract security skills, and building more practical projects.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dcd8da88d00f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>