Tokonomix Council MCP — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Tokonomix multi-model proposals and independent judging: Do disagreement and blind-spot reports remain visible rather than converting consensus into a safety guarantee?

Reviewed Oct 5, 2026 · Documentation review; complete setup workflow not independently tested

Best fit

Multi-model consensus and independent-judge synthesis for high-stakes reasoning.

What we know

MCP that sends a question to multiple frontier models and has a separate judge synthesize a verdict and disagreements. Remote OAuth and local API-key transports have slightly different feature sets.

Maintenance / activity evidence

tokonomix-council-mcp 1.14.0 published 2026-09-28T14:11:45.843Z

Tracking evidence observed Oct 5, 2026. These dated signals do not establish current maintenance. Reviewed enrichment confidence: high.

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Reasoning/model-call authority; no target-system mutation evidenced.
OAuth/API key.
Sensitive access: Potentially sensitive prompts, Account balance/API credential

Primary risks are prompt confidentiality and spend; no external action authority in the council itself.

Full authority, access and evidence limits are available in Review context and Access & risk.

All limitations / Review context

Do disagreement and blind-spot reports remain visible rather than converting consensus into a safety guarantee?

The bounded review of the publisher's documented Tokonomix multi-model proposals and independent judging and release notices identified no disclosed material condition preventing an accurately scoped public tracking entry. The resource's operational permissions remain uncertainty; this assessment does not certify installation safety.

Primary risks are prompt confidentiality and spend; no external action authority in the council itself.
Reasoning/model-call authority; no target-system mutation evidenced.
OAuth/API key.
Prompts are sent to multiple model providers through Tokonomix and synthesized.
Sensitive access: Potentially sensitive prompts, Account balance/API credential
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited.
Documented tool authority and any remote data handling must be checked before installation or use.
Tool inventory, client matrix and permissions completeness were not requirements for this Radar tracking review.
exact local env var name
provider retention details
Every council call spends model balance/credits.
Remote transport lacks some local-only modes/features.
Consensus is not a guarantee of correctness.
Tool calls return 402 when account balance is empty.

Original activity observations

tokonomix-council-mcp 1.14.0 published 2026-09-28T14:11:45.843Z

Technical identity

npm/tokonomix-council-mcp

Exact implementation source: https://www.npmjs.com/package/tokonomix-council-mcp ↗
Reviewed resource type: MCP Server
Package identity: npm:tokonomix-council-mcp
Service endpoint: https://mcp.tokonomix.ai/mcp

Evidence & Sources

Evidence basis: Recorded source evidence
Setup runtime verification: Not independently runtime-tested by MCP100 for this setup workflow.

Reviewed Oct 5, 2026 · Latest source observation Oct 8, 2026

Package registry metadata ↗
tokonomix-council-mcp
Recorded package identity, distribution or download evidence.
Evidence source ↗
tokonomix.ai/en/mcp
Source material retained for this review.
Includes references whose attribution is not established.
Evidence source ↗
tokonomix.ai/en/integrations
Source material retained for this review.
Includes references whose attribution is not established.

Setup · Watchlist

Install & Agents

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Install / connect
Publisher instructions
Publisher-documented package/connection route; replace placeholders and verify host-specific syntax.
Publisher setup example · Publisher instructions
npx -y tokonomix-council-mcp
Configure / authenticate
MCP100 setup guide
Register the package stdio server as publisher documented and configure model/vendor keys that are actually required for selected council backends.
Use correct transport and published host config. Never paste secrets in chat; server registration is separate from software installation.
Environment settings: TOKONOMIX API key for local package (exact env name should be read from package if needed)
Previously reviewed route-specific authentication / prerequisites: Remote OAuth; local package uses Tokonomix API key.
Previously reviewed route-specific authentication / prerequisites: Tokonomix account/balance
Check connection
Suggested check
Confirm council tools are listed; no billable model fan-out as a startup test.
Remove / uninstall / revoke
MCP100 setup guide
If registered as `tokonomix-council-mcp` in Claude Code, remove its exact client entry with `claude mcp remove tokonomix-council-mcp`; for Cursor/Claude Desktop/Codex, remove the matching MCP entry in that host settings. This client removal is MCP100-composed, not a vendor uninstall instruction. Npx does not imply a global install; do not delete user data or revoke credentials implicitly.
Conditional host / package cleanup · Claude Code only · MCP100 composed guide
claude mcp remove tokonomix-council-mcp

Official Documentation & Sources

What it does

MCP that sends a question to multiple frontier models and has a separate judge synthesize a verdict and disagreements. Remote OAuth and local API-key transports have slightly different feature sets.

Capability: AI & Reasoning
Resource type: MCP Server
Best fit: Multi-model consensus and independent-judge synthesis for high-stakes reasoning.

Representative uses

Get multi-model consensus
Surface disagreements/uncertainty
Use independent judge synthesis
Use local full/vision/onboarding extras when needed

Limitations

Every council call spends model balance/credits.
Remote transport lacks some local-only modes/features.
Consensus is not a guarantee of correctness.
Tool calls return 402 when account balance is empty.

Reviewed access

Access & risk

Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.

Authentication: OAuth/API key.
Permissions: Reasoning/model-call authority; no target-system mutation evidenced.
Data handling: Prompts are sent to multiple model providers through Tokonomix and synthesized.
Cost: usage_based
Risk and cautions: Primary risks are prompt confidentiality and spend; no external action authority in the council itself.

Access scope and cautions

Write actions: None in the reviewed inventory
Sensitive access: Potentially sensitive prompts Account balance/API credential
Risk level: moderate

FAQ

Based on reviewed evidence as of 2026-10-08. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Get multi-model consensus; Surface disagreements/uncertainty; Use independent judge synthesis; Use local full/vision/onboarding extras when needed.

What setup and authentication are documented?

Recorded installation/entrypoint step: npx tokonomix-council-mcp. Recorded endpoint: https://mcp.tokonomix.ai/mcp. Recorded transport: stdio, streamable_http. Recorded environment-variable notes: TOKONOMIX API key for local package (exact env name should be read from package if needed). No configuration values are supplied here. Prerequisites: Tokonomix account/balance; MCP client. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: Remote OAuth; local package uses Tokonomix API key.

Can it modify data or perform write operations?

The reviewed inventory lists no write actions. An empty inventory alone does not establish read-only authority. Recorded permission boundary: Reasoning/model-call authority; no target-system mutation evidenced.

What data-handling boundary should I consider?

Prompts are sent to multiple model providers through Tokonomix and synthesized. Recorded sensitive-access context: Potentially sensitive prompts; Account balance/API credential.

What limitations and risks should I check before use?

Primary risks are prompt confidentiality and spend; no external action authority in the council itself. Every council call spends model balance/credits. Remote transport lacks some local-only modes/features. Consensus is not a guarantee of correctness. Tool calls return 402 when account balance is empty. Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited. Documented tool authority and any remote data handling must be checked before installation or use.

What cost model is recorded?

The reviewed cost model is usage based. This classification does not establish current prices, quotas or the cost of every downstream operation. Check the applicable service terms before authorizing paid actions.