# OpenNodes (remote · registry.opennodes.io)

Discover verified AI inference nodes, get the best node+model per task, pre-price chains.

- Trust score: 76/100 (medium)
- Change this week: +2
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `registry.opennodes.io`: 76/100 (this document), [markdown](https://verifymcp.io/servers/opennodes-io-opennodes/registry.md), [page](https://verifymcp.io/servers/opennodes-io-opennodes/registry)
- npm · `@opennodes/cli`: 89/100, [markdown](https://verifymcp.io/servers/opennodes-io-opennodes/opennodes-cli.md), [page](https://verifymcp.io/servers/opennodes-io-opennodes/opennodes-cli)

## Channel facts

- Endpoint: `https://registry.opennodes.io/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.1.14`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 74/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 60/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 2253 tokens (~321/item across 7 items; 7 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 57/100
  - Stability observed for 17 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 7 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 8 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the OpenNodes MCP server?

OpenNodes is a hosted endpoint at https://registry.opennodes.io/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http opennodes-io-opennodes 'https://registry.opennodes.io/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "opennodes-io-opennodes": {
      "url": "https://registry.opennodes.io/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "opennodes-io-opennodes": {
      "type": "http",
      "url": "https://registry.opennodes.io/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.opennodes-io-opennodes]
url = "https://registry.opennodes.io/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "opennodes-io-opennodes": {
      "type": "remote",
      "url": "https://registry.opennodes.io/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add opennodes-io-opennodes --url 'https://registry.opennodes.io/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  opennodes-io-opennodes:
    url: "https://registry.opennodes.io/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "opennodes-io-opennodes": {
      "Transport": "http",
      "Url": "https://registry.opennodes.io/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add opennodes-io-opennodes -t streamable-http -u 'https://registry.opennodes.io/mcp'
```

### Other

```json
{
  "mcpServers": {
    "opennodes-io-opennodes": {
      "type": "http",
      "url": "https://registry.opennodes.io/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-05 (score 76, 0)

- [functional] Server version: 0.1.46 → 0.1.51

### 2026-11-04 (score 76, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: 1854 → 2253
- [functional] Server version: 0.1.42 → 0.1.46
- [functional] New tool “find_residency”
- [cosmetic] “find_locations” added an optional parameter “availability”
- [cosmetic] “find_locations” added an optional parameter “residency”
- [cosmetic] “search_offerings” reworded the description of “region”

### 2026-11-03 (score 76, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “search_offerings” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 242 → 309
- [functional] Server version: 0.1.33 → 0.1.42
- [functional] New tool “find_locations”
- [cosmetic] “search_offerings” added an optional parameter “region”
- [cosmetic] “search_offerings” added an optional parameter “exclude_lineage”
- [cosmetic] “search_offerings” added an optional parameter “local”
- [cosmetic] “search_offerings” added an optional parameter “location_basis”
- [cosmetic] “search_offerings” added an optional parameter “memory_gb”

### 2026-11-02 (score 76, +1)

- [functional] Server version: 0.1.31 → 0.1.33

### 2026-11-01 (score 75, 0)

- [functional] Server version: 0.1.29 → 0.1.31

### 2026-10-31 (score 75, +1)

No change was recorded against any check on this day. Stability & Change Management went from 37 to 40. That category is still filling its 30-day observation window: 11 days of observed history at the previous scan, 12 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-29 (score 74, +1)

- [functional] Server version: 0.1.28 → 0.1.29

### 2026-10-27 (score 73, +1)

- [functional] Server version: 0.1.23 → 0.1.28

## MCP tools (7)

### `recommend` (~256 tokens)

Recommend a node and model

Recommend the best AI node + model for a task, with an enforceable cost estimate and human-readable reasons. Use this first when you know the task but not the model. Pass `task` (sent to this registry, which extracts features and discards the text) or `features` already extracted on your side with extractFeatures from @opennodes/core, which keeps the prompt private. Returns ranked candidates with score components, a cost estimate pinned to the node's card revision, an optional cheaper two-step scenario, and MCP tool categories the task likely needs. To browse instead, use search_offerings; to price a specific chain, use estimate.

Input parameters:

- `est_output_tokens` (integer): Expected answer length in tokens, if known. Improves the cost estimate; otherwise inferred from the task.
- `features` (object): Features from extractFeatures(prompt) in @opennodes/core: task class, size, language and needs, no prompt text. Give this or `task`.
- `policy` (object): Optional constraints and preferences for the recommendation.
- `task` (string): The task description or the prompt itself. Sent to the registry for feature extraction; use `features` to keep it private. Give this or `features`.

### `search_offerings` (~753 tokens)

Search inference offerings

Search AI inference offerings in the OpenNodes registry — registered nodes AND imported catalogs (OpenRouter, Hugging Face Inference Providers, models.dev, local nodes). All filters are optional and combine with AND. Returns { offerings, total, counts }: each offering has node_id, offering_id, model, pricing (USD per MTok), trust tier, serving capabilities, measured probe results, and a rank explanation. Use get_offering for one offering's full card, recommend to let the advisor choose.

Input parameters:

- `exclude_lineage` (string): Comma-separated ISO country codes; drops offerings whose declared model lineage includes any of them, e.g. "CN". Offerings that declare no lineage are kept.
- `family` (string): Model family or exact artifact, e.g. "llama-3.3" or "qwen3". Matches model.family or model.artifact exactly.
- `jurisdiction` (string): Only operators that declare a country inside it: "eea" (EU + Iceland, Liechtenstein, Norway) or comma-separated ISO country codes, e.g. "DE,FI". Registered nodes declare their country; imported listi…
- `lang` (string): Language with minimum grade, "<iso-code>[:basic|strong|native]", e.g. "pl:strong" or "de".
- `limit` (integer): Maximum offerings to return; `total` still counts every match.
- `local` (boolean): true: only models that run on the user's own machine (the Ollama library: no server, free, ranked by Ollama pulls, with local.sizes_memory per size). false: only hosted offerings. Omit for both (host…
- `location_basis` (string): Minimum evidence for the serving locations: claimed < documented (the provider's own dated page) < measured (registry probes) < attested. Offerings without known locations are excluded.
- `max_input_price` (number): Maximum input price in USD per million tokens.
- `memory_gb` (number): RAM/VRAM the user can give a local model, in GB: keeps local models with a size that fits (4-bit estimate). Hosted offerings are unaffected.
- `min_context` (integer): Minimum context window in tokens, e.g. 32768.
- `modality` (string): What the model consumes/produces. Use "text" for chat and completion models.
- `q` (string): Free-text match against the model metadata, offering id, and node id, e.g. "bielik" or "coder".
- `region` (string): Data residency: every pinned region and every live serving site must lie inside this ISO 3166 code, e.g. "CA" (also matches "CA-QC") or "EU" (also matches member states such as "DE"; "EEA" adds IS, L…
- `scheme` (string): Payment scheme the offering must accept. "free" finds no-cost offerings.
- `sort` (string): rank = registry ranking (price, trust, reliability); price = cheapest input first; ttft = fastest first token; tps = highest throughput.
- `supports` (string): Comma-separated features the offering must all support: tool_calls, json_mode, vision, streaming. Example: "tool_calls,json_mode".
- `tier` (string): Minimum trust tier. unverified < community (passed automated checks) < verified (identity and context verified) < attested.

### `get_offering` (~104 tokens)

Get one offering

Full detail for one offering: endpoints, pricing, serving capabilities (context window, supported features, languages, hardware), data policy, trust tier, and the registry's observed probe results. Take node_id and offering_id from search_offerings or recommend results.

Input parameters:

- `node_id` (string, required): Reverse-DNS node id, e.g. "io.opennodes.demo-node".
- `offering_id` (string, required): Offering id within that node, as returned by search_offerings.

### `get_node` (~164 tokens)

Get one node

Who runs a node and how it has behaved: operator, country and regions, endpoints, declared hardware, trust tier and its history (verifications, suspensions, reinstatements, with reasons), recent registry probe results with an availability summary, and its offerings with context window and price. Use it to vet a provider before relying on it; take node_id from search_offerings, recommend or get_offering results. For one offering's full card, use get_offering.

Input parameters:

- `node_id` (string, required): Reverse-DNS node id, e.g. "io.opennodes.demo-node" or "import.openrouter.ai".
- `offering_limit` (integer): Maximum offerings to list; imported catalogs can hold hundreds. `offerings_total` always gives the full count.

### `estimate` (~103 tokens)

Estimate the cost of a chain

Pre-price an invocation chain BEFORE spending money. Each step is either a concrete offering ("node_id/offering_id") or a `select` query that picks the best match, plus expected input and output tokens. Returns per-step cost bounds in USD, the pinned card_revision that makes each estimate enforceable at invocation, and chain totals. Use it to compare routes or budget a multi-model pipeline; nothing is invoked or charged.

Input parameters:

- `steps` (array, required): Steps in execution order.

### `find_locations` (~459 tokens)

Find where models run

Where AI models are served, by country: every known serving site (provider, region, city or facility, live or planned) with its evidence basis (claimed, documented, measured, attested), its source page, a map point (`geo`: lat, lon, precision; `derived` when only the region is known), and the models it serves. Sites whose provider publishes no model list are included with a note. Built from node cards, imported catalogs and curated location seeds. Use it to answer "which providers serve which models in Canada" or "what serves gpt-oss within 500 km of Montréal", or to check a residency requirement before choosing an offering; then search_offerings with `region` to pick one.

Input parameters:

- `availability` (string): With a residency filter: count only commitments offered this widely: default (every customer), opt_in (adds settings and regional endpoints), enterprise (adds contract-only offers; the default).
- `basis` (string): Minimum evidence for a site: claimed < documented < measured < attested.
- `country` (string): ISO 3166 code; a country also matches its subdivisions, e.g. "CA" matches "CA-QC". Omit for every country.
- `exclude_lineage` (string): Comma-separated ISO country codes; drops models whose declared lineage includes any of them, e.g. "CN".
- `model` (string): Substring of a model name or artifact, e.g. "gpt-oss" or "command". Keeps only sites serving a match.
- `near` (string): A point as "lat,lon" in decimal degrees, e.g. "45.50,-73.57" for Montréal. Adds distance_km to each site and sorts nearest first.
- `residency` (string): Only sites covered by a residency commitment at least this strong: processing (inference and stored data stay in the area) or storage (data at rest only).
- `status` (string): Only live or only planned sites. Omit for both.
- `within_km` (number): With `near`: only sites within this great-circle distance. Sites known only to the country are excluded, since their point is a placeholder.

### `find_residency` (~283 tokens)

Find residency commitments

What each AI provider officially promises about where prompts, outputs and stored data stay: per provider (cloud, lab API, inference host or router), its commitments, each with a commitment level (processing: inference and stored data stay in the scope; storage: data at rest only; none; unstated), the region codes it covers, who gets it (default, opt_in, enterprise), the programme name, a short statement and the source page with the date it was checked. Includes routers and providers with no known sites. Use it to answer "which providers keep data in the EU by default" before choosing an offering; then find_locations or search_offerings with `region`.

Input parameters:

- `availability` (string): With a residency filter: count only commitments offered this widely: default (every customer), opt_in (adds settings and regional endpoints), enterprise (adds contract-only offers; the default).
- `commitment` (string): Minimum commitment: processing, or storage (which also matches processing).
- `kind` (string): Only providers of this kind.
- `node_id` (string): Only the provider that covers this registry node id.
- `scope` (string): Region code, e.g. "EU", "DE", "US", "CA". Keeps providers with a processing or storage commitment whose scope lies within it. EU and EEA cover their member states.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/opennodes-io-opennodes/registry#diagnostics

## Score history

- 2026-11-05: 76
- 2026-11-04: 76
- 2026-11-03: 76
- 2026-11-02: 76
- 2026-11-01: 75
- 2026-10-31: 75
- 2026-10-30: 74
- 2026-10-29: 74
- 2026-10-28: 73
- 2026-10-27: 73
- 2026-10-26: 72
- 2026-10-25: 72
- 2026-10-24: 72
- 2026-10-23: 71
- 2026-10-22: 71
- 2026-10-21: 70
- 2026-10-20: 70
- 2026-10-19: 69

## Common questions

### What is the OpenNodes MCP server?

OpenNodes is an MCP server listed in the public MCP registry as io.github.opennodes-io/opennodes. Discover verified AI inference nodes, get the best node+model per task, pre-price chains. This page covers its hosted endpoint (https://registry.opennodes.io/mcp).

### Is the OpenNodes MCP server safe to use?

OpenNodes scores 76 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the OpenNodes MCP server expose?

OpenNodes exposes 7 tools: recommend, search_offerings, get_offering, get_node, estimate, and 2 more. Their descriptions and schemas cost roughly 2,122 tokens of context every time the server is loaded.

### Does the OpenNodes MCP server require authentication?

No. We connected to OpenNodes without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the OpenNodes MCP server still maintained?

OpenNodes is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://registry.opennodes.io/mcp
- Repository: https://github.com/opennodes-io/opennodes
- Website: https://opennodes.io/
- Changelog RSS feed: https://verifymcp.io/servers/opennodes-io-opennodes/registry.xml
- Changelog JSON feed: https://verifymcp.io/servers/opennodes-io-opennodes/registry.json
- HTML version of this page: https://verifymcp.io/servers/opennodes-io-opennodes/registry