Devpost
Participate in our public hackathons
Devpost for Teams
Access your company's private hackathons
Grow your developer ecosystem and promote your platform
Drive innovation, collaboration, and retention within your organization
By use case
Blog
Insights into hackathon planning and participation
Customer stories
Inspiration from peers and other industry leaders
Planning guides
Best practices for planning online and in-person hackathons
Webinars & events
Upcoming events and on-demand recordings
Help desk
Common questions and support documentation
Sort by:
Evidence-based reconciliation for incoming customer payments.
Lauudit, the trust layer for legal AI: a Chrome extension that verifies cases, citations and whether judgments support AI-generated claims, giving users source-linked checks before they rely on them.
Evidence-grounded recovery support for concussion: short cognitive, visual and movement assessments combined with personal baselines, longitudinal trends, and safe AI explanations.
Turns "idk what do you want to do" into one agreed plan — three taps each, twenty seconds, no group chat spiral.
WebMCPify is a domain-agnostic CLI and local MCP workflow where a human and a coding agent discover, draft, review, apply, test, and evaluate WebMCP integrations.
A browser-based XTT plate-analysis app that delivers exploratory image-derived metabolic-activity insights with transparent QC, mapped controls, and WebMCP-assisted setup.
A real-time strategy game that keeps more depth than one player can manage, because your agent playsalongside you through its own interface instead of your screen.
Blind Spot uses AI and the Feynman Technique to evaluate your true understanding of any topic. It identifies hidden knowledge gaps and cures the "Illusion of Competence" to ensure true mastery.
WebMCP tooling for FullCalendar React applications without replacing the UI or the backend.
Maps that agents can understand
Widgets for agents. Your agent drafts into the designer you are looking at over WebMCP, you save, and ChatGPT, Claude and Copilot render it inline over MCP.
Call the play. Lock your vision. Let AI replan the rest.
An agent-native semantic workspace where ChatGPT proposes validated visual structures through WebMCP—and the human retains authority to accept, export, or undo.
Train Platform: Agentic cinema post-production OS powered by Google Cloud & Gemini 2.0. Orchestrates editorial, VFX, sound & color with autonomous AI sentries and OpenTimelineIO export and Grafana
Fold a shared 3D sheet with an agent that reads geometry, not pixels.
WebMCP lets pages hand AI agents real power. This shows how one malicious support ticket weaponizes it and which defenses actually stop it. Attack, toggle, re-run, compare.
Ask an agent to build a slide, or edit it by hand — nothing regenerates. You're both on the same live deck.
This demo explores user interfaces that adapt to the humans that browse it, and asks what would the web be if you could B-Y-O-U-I: Bring your own user interface.
WebMCP tells an agent what a site can do. Nothing tells it what a site can take back. Concord is that missing layer: it computes what it can honestly promise across sites before contacting any.
AI builds websites fast and generic. BlockSmith captures a design system from any site in seconds, hands the rules to the AI, checks its work, and tells it no when it breaks them.
The ecosystem quotes "10x faster" for WebMCP with no methodology behind it. AgentPerf measures it: 2.4x on a small page, 13.8x on a real one. And it ships the React layer that produces the gap.
PlanOnIt gives a WebMCP agent real tools to plan your evening out — film, transport, dinner — while you keep the final approval.
Dungeon Familiar: A Co-op Game You Play With an AI Agent
Point at what looks wrong. Say four words. Get a measured answer, and a safe fix.
