Architecture-first Python scaffold for an auditable multi-agent corporate credit desk using A2A, MCP, deterministic credit policies, model routing, and OpenTelemetry.
Read-only linter for CLAUDE.md, AGENTS.md, SKILL.md and agent rules files: vendor limits, tokens loaded per session, hidden instructions and invisible characters, and files Claude Code or Codex never read. Claude Code plugin, MCP server, CLI, GitHub Action and pre-commit hook. No network. MIT.
Kai is an AI-powered investment analysis system that uses a multi-agent debate architecture to provide explainable, trustworthy, and data-driven investment insights.
Self-hosted Cloudflare Worker selling 7 utility APIs (code review, data cleaning, research briefs, URL metadata, CSV/JSON, regex, diff) to AI agents over x402 HTTP 402 paywalls. No account, no API key.
This project is a technical talk demonstration showcasing A2A (Agent-to-Agent) communication connecting multiple distinct AI agent frameworks and MCP servers together under a unified orchestration layer.
Multi-agent collaboration hub: turn opencode, claude code and other AI coding tools into a swarm — dispatch tasks across workspaces via standard MCP tools, watch and answer permission prompts live from the web console, and connect through Feishu. Self-hosted with a single service + SQLite.