Multi-Agent Frameworks

A JAX-powered agent-based model turns households, banks, markets, and policy rules into one compiled simulation, then adds a world model so the economy can be learned as well as run.

A multimodal civic stack for Bengaluru that cross-checks text, voice, and images before it ever decides where a report should go.

A deliberately simple terminal transport for Claude and Codex, with transcript verification, pane discovery, and just enough structure to keep multi-agent work from drifting off the rails.

A command-line agency that researches prospects, scores opportunities, and generates client-ready deliverables by combining parallel agents, weighted rubrics, and a PDF synthesis layer.

Why the future of AI coding belongs to "Fresh-Instance" pipelines and adversarial review loops rather than long-form chat.

A low-code LiveKit builder that turns handoffs, tasks, and dispatch flows into executable Python, then runs them in a hot-reloading preview runtime.

A modular skill system that splits ad strategy into parallel specialist agents, scores readiness across audience, creative, funnel, budget, and competition, and packages the result into a polished report.

A single-file open-source manifesto that borrows Agile’s structure, treats cognitive bandwidth as the new unit of work, and reframes AI-native development as orchestration, not code entry.

How a modular multi-agent system uses "few-shot" orchestration to bridge the gap between general AI and specialized Web3 infrastructure.

How a zero-cost GitHub Actions pipeline uses a swarm of self-critiquing agents to run a Berlin culture magazine without a single human editor.

By pitting Claude and GPT against each other in a parallel race for root, this framework replaces the lone AI hacker with a ruthless, self-coordinating team.

A look at how Appwrite combines stateless turns, just-in-time skill loading, and UI protocols to turn natural language into safe, structured infrastructure actions.

How a Django and PyTorch pipeline aligns mismatched photos, ignores vegetation noise, and lets humans verify the damage AI thinks it sees.

This Socket.io-driven ATS does more than rank candidates. It evaluates job posts, scores fit across multiple dimensions, and pushes recruiting toward a real-time conversation.

A meta-skill that turns OpenCode into an agent-making system, with schemas, workflows, permission patterns, and self-audits instead of ad hoc prompts.

A Homebrew tap for Gas City, an orchestration SDK that pairs terminal sessions, work tracking, and Dolt-backed history to make multi-agent systems inspectable, repeatable, and less of a black box.

A Python middleware project that routes carbon calculations through MCP, the OpenAI Apps SDK, and a minimal fallback server, then returns either plain text or a live widget.

How a Next.js application rejected the cloud to enable real-time, optimistic co-editing between humans and autonomous agents.

Moving beyond the everything-app prompt to a registry of specialized micro-agents and verified workflows.

How a YAML-to-XML pipeline is turning 'vibe coding' into a disciplined agentic factory.

A two-agent Python pipeline that turns GitHub PRs into structured AI reviews, then mirrors approved file changes across repositories without a local clone.

A modular MOT pipeline that keeps IDs stable through occlusion, rewrites lost identities with appearance embeddings, and extends the same logic across two cameras.

A LangGraph pipeline that routes topics by epistemic mode, fans out section drafts in parallel, and patches visuals into the final article only after the argument is clear.

A deep look at a fail-safe agent stack that chooses between search, summary, or both, routes tasks across multiple models, and keeps the whole system observable and recoverable.

A measured triage system that uses LLMs for judgment, then uses deterministic rules, safe execution, and human approval to keep the judgment from going off the rails.

A Spring Boot HRIS with an MCP bridge, department-scoped JWT claims, and guardrails that make agentic HR workflows feel less like a demo and more like an operating model.

A Spring Boot and React complaint system that uses an LLM for triage, suggestions, and live escalation, then wraps it in JWT auth, rate limits, and real-time notifications.

A real-time DRL trading system that turns machine signals into pending trades, pushes state through Redis and WebSockets, and lets humans veto the machine before money moves.

A multimodal grievance platform pairs LangGraph agents, PostGIS routing, and AI verification to turn municipal complaints into a traceable, spatially aware workflow.

A brutally honest AI validator that combines Claude, local geography, and market signals to tell founders what their idea is up against before they build it.

A deep dive into the open-source triage system that normalizes, validates, audits, and escalates before an LLM can turn a message into work.

A lean Node, MySQL, and vanilla JS stack coordinates judge, counsel, and advisor agents into a structured legal debate system that remembers, scores, and teaches.

A deep dive into the agentic system that turns agricultural waste into actionable recommendations by combining clarifying questions, hybrid retrieval, trusted market data, and deterministic ROI calculations.

A tiny repository, a big ambition. The `.gitignore` hints at a multi-agent support system built for retrieval, compliance, and action, not just conversation.

A deep dive into a multi-agent system that turns marketing goals into a planned sequence of research, strategy, and creative execution without letting the context window balloon out of control.

A static site, a skill bundle, and a CROPS framework live in one source tree, so the same design process can coach people and LLMs.

edge-cityverse is only a README, but the name points to something bigger: an edge-native layer for city-scale coordination, where local nodes keep the working state and shared rules hold the network together.

How a collection of hollow shell scripts and aggressive credential injection turns an empty repository into an autonomous, multi-agent development team.

How a modular Node.js orchestrator uses git worktrees and flat files to safely manage heterogeneous agent swarms without the bloat of modern frameworks.

How a modular microservices architecture uses Kubernetes, Kali Linux, and the Model Context Protocol to give AI swarms a safe place to execute complex tasks.

How a React application uses VNC streams and Kubernetes namespaces to build a shared, ephemeral workspace for security researchers and autonomous agents.

How a pure Swift architecture turns local LLMs into a drag-and-drop playground for agent orchestration.

How a multi-agent pipeline uses Google ADK to bring peer review, quality assurance, and empathy to automated grading.

How a native macOS shell leverages Ghostty's rendering engine to turn the traditional terminal into an orchestration dashboard for autonomous code.

How a bleeding-edge reference architecture uses the Actor model to replace event-driven boilerplate with fully typed, synchronized shared state.

How Manaflow AI combined Ghostty's rendering engine and Mac-native Swift to orchestrate swarms of parallel coding agents.

How an experimental Go orchestrator uses event sourcing and RPG mechanics to solve the AI context collapse problem.

Moving beyond the chat box to a multi-agent factory floor where AI personas plan, architect, and execute in lockstep.