REA
Verified · 33 days agoReverse engineer anything from your terminal or agent with one CLI and MCP server.
Multi-model consensus: 2-6 frontier LLMs answer, an independent judge synthesises one answer.
claude mcp add council -- npx -y [email protected]
Thin MCP client for tokonomix.ai, a paid EU-hosted gateway that fans a question out to 2-6 cross-vendor frontier models in parallel and has a disjoint 'judge' model synthesize one answer while preserving disagreements. Ten tokonomix_-prefixed tools (consensus ask, single-model passthrough, model catalog, balance, keyless email onboarding; two are documented as dormant server-side), TypeScript, MIT, npx install, everything metered through a tok_live_ API key with €5 free credit. Unusually, its self-published benchmarks concede the council ties the best single model on accuracy and showed no net bug-catch uplift on SWE-bench — the pitch is variance-elimination and auditability, not correctness. It is a solo-maintainer, zero-star project one month old, and the ~1,700 npm downloads/week are implausible as organic use given 25 releases in that month, so read them as mirror noise. For teams wanting a second-opinion layer on high-stakes agent decisions with per-key spend caps, who accept a brand-new metered SaaS in the loop.
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Agent teams, swarms, memory and hooks for Claude Code and Codex, served over MCP.
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Draw and visually collaborate with your agents on tldraw's canvas.
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Keyword research, site audits, rank tracking and AI visibility for your AI agent.