Verlon MCP — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Verlon infrastructure gates, logs and opt-in model switching: Does the write flag remain required for switch_model as the planned mutation surface grows?

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

Best fit

Configure and inspect LLM routing, quality/cost tracking, experiments and runtime gates from an MCP client.

What we know

MCP surface for Verlon's provider-neutral LLM gateway, experiments, logs/recommendations and control-plane configuration.

Maintenance / activity evidence

@verlon-ai/mcp 0.4.1 published 2026-09-13T05:04:45.501Z

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.
Observability plus routing/experiment/gate configuration. Writes: Create experiments; Change model routing; Configure gates/recommendations.
Verlon account credential.
Sensitive access: Prompt/response or telemetry depending on gateway policy, Provider routing, Cost/quality data

Changing production routing/experiments can affect quality, cost and behavior.

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

All limitations / Review context

Does the write flag remain required for switch_model as the planned mutation surface grows?

The bounded review of the publisher's documented Verlon infrastructure gates, logs and opt-in model switching 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.

Changing production routing/experiments can affect quality, cost and behavior.
Observability plus routing/experiment/gate configuration.
Verlon account credential.
LLM requests/telemetry pass through or are observed by Verlon depending on gateway configuration.
Reviewed write actions: Create experiments; Change model routing; Configure gates/recommendations
Sensitive access: Prompt/response or telemetry depending on gateway policy, Provider routing, Cost/quality data
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 API-key env
retention/privacy specifics
public repository identity
Control-plane writes can alter production model behavior and spending.
The platform supports many upstream providers and acts as a gateway.
Exact API-key/env setup and retention policy were not captured here.

Original activity observations

@verlon-ai/mcp 0.4.1 published 2026-09-13T05:04:45.501Z

Technical identity

npm/@verlon-ai/mcp

Exact implementation source: https://www.npmjs.com/package/@verlon-ai/mcp ↗
Reviewed resource type: MCP Server
Package identity: npm:@verlon-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 ↗
@verlon-ai/mcp
Recorded package identity, distribution or download evidence.
Evidence source ↗
verlon.ai/
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
Read-only tools by default; `switch_model` only if --enable-writes is used.
publisher documented install or connect · Publisher instructions
npx -y @verlon-ai/mcp
Configure / authenticate
MCP100 setup guide
VERLON_API_KEY required for account tools; a read-only model list is safer than production gateway policy changes.
Environment variables referenced by this route (respect the alternatives above): VERLON_API_KEY
Check connection
Suggested check
User-side suggested check (not executed): inspect registered tools/connection and perform publisher-supported read-only status/list operations. Do not send, purchase, publish, delete, sign, trade or alter data as a smoke test.
Remove / uninstall / revoke
MCP100 setup guide
Generic client guidance (not publisher-specific uninstall): unregister MCP entry in the chosen client. If installed globally, remove the npm package separately. Revoke provider access separately through account controls, where supported. This does not erase local data or revoke funds/credentials automatically.
Unregister the named MCP entry from the actual host separately from removing a global npm install; no claim that either revokes remote credentials.
Remove local stored keys/tokens and revoke or rotate corresponding provider authorization using documented provider account controls if available; exact procedure not verified for this package. Wallet private keys cannot be revoked.

Official Documentation & Sources

What it does

MCP surface for Verlon's provider-neutral LLM gateway, experiments, logs/recommendations and control-plane configuration.

Capability: AI & Reasoning
Resource type: MCP Server
Best fit: Configure and inspect LLM routing, quality/cost tracking, experiments and runtime gates from an MCP client.

Representative uses

Inspect live LLM traffic/cost/quality
Create experiments
Change routing/model choices
Configure runtime gates/recommendations

Limitations

Control-plane writes can alter production model behavior and spending.
The platform supports many upstream providers and acts as a gateway.
Exact API-key/env setup and retention policy were not captured here.

Reviewed access

Access & risk

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

Authentication: Verlon account credential.
Permissions: Observability plus routing/experiment/gate configuration.
Data handling: LLM requests/telemetry pass through or are observed by Verlon depending on gateway configuration.
Cost: plan_based
Risk and cautions: Changing production routing/experiments can affect quality, cost and behavior.

Access scope and cautions

Write actions: Create experiments Change model routing Configure gates/recommendations
Sensitive access: Prompt/response or telemetry depending on gateway policy Provider routing Cost/quality data
Risk level: high

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: Inspect live LLM traffic/cost/quality; Create experiments; Change routing/model choices; Configure runtime gates/recommendations.

What authentication is documented?

Verlon account credential.

Can it modify data or perform write operations?

Documented write actions include: Create experiments; Change model routing; Configure gates/recommendations. Recorded permission boundary: Observability plus routing/experiment/gate configuration.

What data-handling boundary should I consider?

LLM requests/telemetry pass through or are observed by Verlon depending on gateway configuration. Recorded sensitive-access context: Prompt/response or telemetry depending on gateway policy; Provider routing; Cost/quality data.

What limitations and risks should I check before use?

Changing production routing/experiments can affect quality, cost and behavior. Control-plane writes can alter production model behavior and spending. The platform supports many upstream providers and acts as a gateway. Exact API-key/env setup and retention policy were not captured here. 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 plan 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.