The latest articles on DEV Community by hesham (@hesham-mansy). https://dev.to/hesham-mansy https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4220099%2F4ac6528d-4743-4746-a79d-46f09fb101f3.png https://dev.to/hesham-mansy en hesham Sun, 01 Nov 2026 06:47:02 +0000 https://dev.to/hesham-mansy/i-built-intentchord-to-bring-human-control-to-ai-coding-agents-contributors-welcome-5d7h https://dev.to/hesham-mansy/i-built-intentchord-to-bring-human-control-to-ai-coding-agents-contributors-welcome-5d7h <p>AI coding agents can write code, modify files, and report that a task is complete.</p> <p>But a successful-looking response doesn't necessarily answer some important questions:</p> <ul> <li>Was the agent working on the correct source?</li> <li>Was the change within the approved scope?</li> <li>Is the reported result supported by actual evidence?</li> <li>Who decides whether the work is accepted?</li> </ul> <p>These questions led me to build <strong>IntentChord</strong>.</p> <h2> What is IntentChord? </h2> <p>IntentChord is a local-first, human-governed development workflow that connects an owner's intent to workspace context, confirmed scope, agent-specific execution contracts, evidence-based review, and a final human decision.</p> <p>Its central principle is simple:</p> <p><strong>An AI agent's report is not the same as verified execution evidence, and an agent's claim of success is not the owner's approval.</strong></p> <h2> What's in the project? </h2> <p>The repository includes:</p> <ul> <li>Task and scope authority management</li> <li>Workspace discovery, indexing, and source-context foundations</li> <li>Governed execution contracts for Codex, Claude Code, and Gemini CLI</li> <li>Execution evidence and report-verification logic</li> <li>Human review and decision workflows</li> <li>A Next.js, React, TypeScript, and PostgreSQL-based application</li> </ul> <p>The project is still evolving and is not presented as a production-certified system.</p> <h2> I'm opening the project to contributors </h2> <p>I've made the source public because I'd like to develop IntentChord with other developers.</p> <p>I'm particularly interested in contributions involving code quality, testing, developer onboarding, architecture discussions, and improvements to the human-in-the-loop workflow.</p> <p>There are already beginner-friendly and documentation tasks available in GitHub Issues.</p> <p><strong>Repository:</strong> <a href="https://github.com/hesham-m/IntentChord" rel="noopener noreferrer">https://github.com/hesham-m/IntentChord</a></p> <p><strong>Contribution opportunities:</strong> <a href="https://github.com/hesham-m/IntentChord/issues" rel="noopener noreferrer">https://github.com/hesham-m/IntentChord/issues</a></p> <p>The project is source-available under <strong>PolyForm Noncommercial 1.0.0</strong>, rather than an OSI-approved open-source license. Please review the repository's license and contribution guidelines before contributing.</p> <p>I'd be interested to hear how other developers handle agent authority, scope changes, and evidence verification in their AI-assisted development workflows.</p> <p><strong>What would you improve or challenge in this approach?</strong></p> ai typescript github discuss