Anthropic

Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

Anthropic

Anthropic’s Constitutional AI training approach research focuses on developing AI systems safe by design and aligned with human values. By prioritizing safety, we can create strong and corrigible AI systems that are safe for humans to use.

General
Company	Anthropic
Founded	2021
Discord	https://discord.gg/lablab-ai-877056448956346408

Anthropic Claude

Claude is your friendly and versatile AI language model that can assist you as a company representative, research assistant, creative partner, or task automator.

Claude is Safe, Clever, and Yours. Built with safety at its core and with industry leading security practices, Claude can be customized to handle complex multi-step instructions and help you achieve your tasks.

You can easily use Claude for your app, and all necessary APIs, boilerplates, tutorials explaining how to do so and more, you can find on our Claude tech page.

Claude Code

Claude Code is a command-line tool from Anthropic for agentic coding. It enables Claude to refactor, debug, and manage code directly in the terminal. You can find more information on our Claude Code tech page.

Edit on GitHub

Anthropic AI Technologies Hackathon projects

Discover innovative solutions crafted with Anthropic AI Technologies, developed by our community members during our engaging hackathons.

AMD2_PKK

``` /input/tasks.json [{task_id, prompt}] <- NO category field | v classify rules (0 tokens, ~0 ms) for the confident majority; | the local model decides the ambiguous tail (also 0 tokens) v budget reads N, MEASURES ITS OWN tok/s, picks a depth: | FULL / SINGLE / FAST / REMOTE v solve math -> model writes Python -> EXECUTE -> exact answer | logic -> model writes an enumeration -> EXECUTE -> filter | code_gen -> generate -> must import and run -> repair | code_debug -> name the bug + corrected code -> must run | sentiment -> label + justification (the guide requires both) | ner -> typed entity lines | summarise -> length constraint enforced after generation | factual -> direct answer (the only unverifiable category) v accept verified by execution? keep it. otherwise mean-token-logprob | risk vs a per-category threshold v escalate ONLY on: clock exhaustion, local failure, or (opt-in) low confidence v /output/results.json [{task_id, answer}] ```

ConsultIn

Quantivo AI, also known as BOA (Business Opportunity Analysis), is an AI-powered SaaS platform that helps small business owners and entrepreneurs make data-driven decisions about their business opportunities. Given a business's context — category, location, growth stage, and goals — the system automatically scrapes real local market data, routes and filters it through a deterministic classification pipeline, and runs parallel sentiment and SWOT analysis using LLM agents orchestrated via LangGraph. The result is a comprehensive report featuring an executive summary, market insights, actionable recommendations, and a visual heatmap of the local competitive landscape. The platform is built around a contract-first architecture: a shared Pydantic schema and Protocol-based interface layer let independent workstreams — scraping, routing, retrieval, agent reasoning, and orchestration — develop and test against mocks in parallel before wiring in production components. Under the hood, Quantivo AI uses a hybrid dense-and-sparse retrieval system (Qdrant with BGE-M3 embeddings) fused via reciprocal rank fusion, and every confidence score is computed quantitatively from source count, agreement, and recency rather than left to subjective LLM judgment. The system is designed for graceful degradation: if any single data source or agent fails, the pipeline still produces a partial, clearly-labeled report instead of failing outright. Quantivo AI was built for the AMD Developer Hackathon (ACT II, Track Unicorn), with LLM inference and embedding generation running on AMD MI300X/MI350 GPU hardware via Fireworks AI and a self-hosted BGE-M3 embedding server, demonstrating a production-realistic AI pipeline built entirely on AMD's AI stack.

ConsultIn

AuraOS

Aura is a sovereign, neuro-symbolic AI operating system designed to make advanced artificial intelligence more transparent, efficient, collaborative, and accountable. Instead of sending every instruction directly to a language model, Aura can locally interpret natural-language intent, organize it into structured semantic and routing representations, identify relevant code or knowledge, and prepare a bounded task before model use is required. Aura combines deterministic routing, a 4,096-primitive lexical system, polysynthetic six-slot intent structures, visual code topology, guarded human-agent workflows, and a Learning Arena where verified experiences can be tested before becoming reusable knowledge. Its Observatory allows users to see how Aura understood an instruction, while the Human Agent Arena supports node-specific dialogue, approval gates, verification, and preserved attempt artifacts—including failed work that may help future refactoring. Aura is built around human authority. Models can propose, explain, investigate, and assist, but they cannot silently expand scope, bypass evidence requirements, or merge changes without approval. The result is an AI architecture focused not only on capability, but on sovereignty, inspectability, learning, and responsible action.

Narraflow

NarraFlow is a multi-agent AI storytelling platform that breaks story generation into specialized stages instead of relying on one model. A Director Agent takes a user prompt and builds a blueprint (title, genre, tone, pacing, key events), and a Writer Agent turns that blueprint into full scenes, dialogue, and narration. This modular setup makes the system extensible — future agents could handle character consistency, world-building, memory, editing, illustrations, voice narration, translation, and interactivity, all without needing to redesign the core system. The Director Agent is kept lightweight for efficiency, keeping generation fast while still giving reliable structure to guide the Writer Agent. Potential uses span creative writing, education, game dialogue, screenplay drafts, role-playing campaigns, and personalized or AI-assisted content creation more broadly. Long-term, the goal is a fuller storytelling ecosystem with persistent character memory, illustrations, branching plots, multilingual support, collaborative writing, and stronger consistency checks — using this multi-agent, structured-planning approach as a template for how modular AI systems can scale creative tasks better than a single monolithic model

TRANSDUCER - Style-Aware Video Captioning-Agent

TRANSDUCER is an autonomous Track-2 agent that watches a video and retells it in four distinct voices: formal, sarcastic, humorous_tech, and humorous_non_tech. Instead of brittle rule-lists, each style is a character — a precise observer, a world-weary wit, a burnt-out engineer, a bewildered dad — handed the actual video frames so the humor reacts to what is genuinely on screen and never invents details the LLM-judge would penalize. Per clip (~15–25s): the download is streamed and salvaged even when truncated, four keyframes are sampled across the whole clip at 1024px, and a single multimodal call to Kimi K2.6 on Fireworks AI returns all four captions at once — forcing sharp contrast between voices — followed by per-style retries, a deterministic fallback ladder guaranteeing complete valid output, and an artifact sanitizer. Engineered for the real grading box (4GB RAM, 2 vCPU, keyless, cold anonymous pull): three clips run concurrently, every clip finishes under 30 seconds, and no style is ever missing. Across 20+ leaderboard-scored builds — one change at a time, kept or reverted on the number — it climbed 0.61 → 0.84 → 0.91, peaking at 0.92, top-3 on the hidden set.