Research & Data

Ten AI agents deliberate every prediction-market trade. Restraint is anchored on-chain. Polymarket CLOB + Arc Testnet. 11 LLM providers, one env var. Half-Kelly with caps, self-calibrating, fully open-source.

OpenClaw skill for AI-assisted interactive learning: spaced repetition, active recall, Feynman technique, interleaving & deliberate practice. OpenClaw技能：AI辅助的互动学习协议，整合间隔重复、主动回忆、费曼学习法、交错练习和精细加工提问。

📡 Antenna — Inter-host OpenClaw messaging over Tailscale

Local DevTools/F12 evidence runtime for AI agents and AppSec workflows.

Autonomous AI Agent Playbooks. - Agent harnessing and orchestration for complex, repeatable, verifiable workflows.

Composable pi coding agent with MCP, LSP, agent chains, prompt presets, and local eval telemetry

Multitenant backend that lets agents discover and execute against any HTTP API. MCP-native — exposes search and execute tools via the Model Context Protocol. Connectors for GitHub, Jenkins, GitLab, Jira, Kubernetes, Linear, and Slack, plus user-supplied OpenAPI/bare-URL connections.

Persistent memory + coordination layer for AI agents. Local SQLite, vector recall, MCP server. Mnemo Connect turns one hub into the dispatcher for a distributed agent team across N machines.

Cortex - AI Agent 的第二大脑。单二进制本地知识库，MCP 原生支持，混合搜索 + Agent 记忆系统。

AI-orchestration kanban board — a co-working control plane for humans and AI coding agents. Next.js 16, React 19, xterm.js + node-pty, MCP.

High-performance MCP server for real-time multi-source enterprise data delivery to LLMs. Entity-based data virtualization

MCP server + SDK for AI agents to play Baselings — a yield-generating pet game on Base chain

Give your AI agent a real browser identity. MCP server with persistent personas, anti-detection browser, and credential vault.

AI scouting agent for World Cup 2026 — find hidden gems using natural language queries powered by Gemini, MongoDB Atlas Vector Search & MCP

Demo of AI agents operating enterprise software through browser control via Playwright MCP — bring your own site, the agent operates it

Connect AI agents to any API, anywhere. Securely. Open-source gateway that proxies requests, injects credentials automatically, punches through NAT to reach localhost services, and wraps REST APIs as MCP tools. Per-agent isolation. Never expose a raw key.

Offline speech-to-text & speaker diarization for AI agents — Whisper ASR + MCP server + CLI + Python bindings, fully on-device, no cloud | 离线语音识别与说话人分离,Agent 开箱即用

Agent Receipts — cryptographically signed audit trails for AI agent actions. Protocol spec, SDKs (Go, TypeScript, Python), and MCP proxy.

AI-powered CLI for structured repository audits with Codex or Claude Code, including evidence packs, risk findings, roadmap reports and parallel analysis.

Complete MCP server exposing all Hacker News Tool operations. Zero config, pre-built workflows

A harness that knows your project — and stays that way. Interview-shaped, grade-gated, self-evolving AI coding harness for Claude Code · Cursor · Codex. Project profiler + 10-dim interview, auto-fix grade-gated review, block-merge edit preservation, anti-rot crawlers, worktree isolation.

Cross-agent hook manager for AI coding assistants — installs and orchestrates lifecycle hooks across Claude Code, Cursor, and Gemini from a single manifest.

Claude Code plugin: warmup for native /goal + autoresearch (deterministic keep-or-revert metric loop).

OpenClaw skill for paid LDM Inbox Check MCP placement tests before outbound campaigns