Multi-model consensus: 2-6 frontier LLMs answer, an independent judge synthesises one answer.
Tokonomix Council MCP server exists for a simple reason — assistants are far more useful when they can act on Tokonomix Council directly instead of describing what you should do. Multi-model consensus: 2-6 frontier LLMs answer, an independent judge synthesises one answer.
Council is one specialized component in a larger stack. Each owns exactly one responsibility, so they compose cleanly:
Being a remote server, there is no local install. You register the endpoint with your client, authorise it once, and the tools appear.
Once Tokonomix Council is connected, these are the calls the assistant has available:
Independence — proposers answer blind and in parallel; the judge is disjoint from them and cross-family. No model grades its own workAuditability — every decision returns who proposed, who judged, and what it cost, so it can be reconstructedTransparency — what was not settled is surfaced (the blind-spots section, the diff map), never hidden behind one answerReproducibility — the same input + method re-runs to the same finding; record the models and date, because the default council can changeTraceability — a request_id and a billing breakdown tie a stored decision back to exactly how it was producedProposers — the 2–6 models that answer the prompt. Chosen for cross-family decorrelation, not just raw score. Empirically selected from Tokonomix's ownJudge — disjoint from the proposers by default, cross-family, never scoring its own answer. Budget is auto-sized to the proposer countGateway — runs the orchestration, applies region/residency policy, meters cost, returns one result with a billing breakdown and a request_id. The gatewayPer — key policy** — budget caps, default council/route, and allow-lists are set on the API key, not re-decided per call, so an unattended run can'tCapability — StatusAmong the payments and commerce options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Tokonomix Council's toolset — Independence, Auditability, Transparency and 7 more — is a fair guide to whether it matches your workflow. It is maintained by ai.tokonomix; worth a glance at recent repository activity before you build anything load-bearing on it.
SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.
| Tool | What it does |
|---|---|
| Independence | proposers answer blind and in parallel; the judge is disjoint from them and cross-family. No model grades its own work. |
| Auditability | every decision returns who proposed, who judged, and what it cost, so it can be reconstructed. |
| Transparency | what was *not* settled is surfaced (the blind-spots section, the diff map), never hidden behind one answer. |
| Reproducibility | the same input + method re-runs to the same finding; record the models and date, because the default council can change. |
| Traceability | a request_id and a billing breakdown tie a stored decision back to exactly how it was produced. |
| Proposers | the 2–6 models that answer the prompt. Chosen for cross-family decorrelation, not just raw score. Empirically selected from Tokonomix's own measurement layer (intelligence tests, arena games via TrueSkill, judge-reputati |
| Judge | disjoint from the proposers by default, cross-family, never scoring its own answer. Budget is auto-sized to the proposer count. |
| Gateway | runs the orchestration, applies region/residency policy, meters cost, returns one result with a billing breakdown and a request_id. The gateway itself is EU-hosted; region selection controls which *model providers* see y |
| Per | key policy** — budget caps, default council/route, and allow-lists are set *on the API key*, not re-decided per call, so an unattended run can't overspen |
| Capability | Status |
Payments infrastructure meets AI — customers, invoices, subscriptions and current API docs from Stripe's server.
Build for Shopify with current APIs — the official Dev MCP server for schema-accurate storefront and admin code.
Give your AI assistant a supervised window into your Interactive Brokers account — balances, positions, live quotes and orders.
Invoices, payments and disputes through PayPal's official server — commerce operations by conversation.
Enables querying and listing Azure resources and costs directly from an MCP client.
Empowers LLMs with Bitcoin Lightning Network payment capabilities via the ZBD API.