# Nonobench (remote · www.nonobench.com)

An open-source benchmark of how well LLMs solve nonogram puzzles, from 5x5 to 20x20.

- Trust score: 86/100 (high trust)
- Change this week: 0
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `www.nonobench.com`: 86/100 (this document), [markdown](https://verifymcp.io/servers/mauricekleine-nonobench/www.md), [page](https://verifymcp.io/servers/mauricekleine-nonobench/www)

## Channel facts

- Endpoint: `https://www.nonobench.com/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.1.0`

## 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**: 78/100
  - AI-judged instruction clarity (excellent).
  - Tool/resource definitions use about 956 tokens (~86/item across 11 items; 11 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 100/100
  - No destabilizing schema changes in the last 30 days.
- **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.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 11 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 12 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

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

Nonobench is a hosted endpoint at https://www.nonobench.com/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 mauricekleine-nonobench 'https://www.nonobench.com/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "mauricekleine-nonobench": {
      "url": "https://www.nonobench.com/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "mauricekleine-nonobench": {
      "type": "http",
      "url": "https://www.nonobench.com/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.mauricekleine-nonobench]
url = "https://www.nonobench.com/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "mauricekleine-nonobench": {
      "type": "remote",
      "url": "https://www.nonobench.com/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add mauricekleine-nonobench --url 'https://www.nonobench.com/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  mauricekleine-nonobench:
    url: "https://www.nonobench.com/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "mauricekleine-nonobench": {
      "Transport": "http",
      "Url": "https://www.nonobench.com/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add mauricekleine-nonobench -t streamable-http -u 'https://www.nonobench.com/mcp'
```

### Other

```json
{
  "mcpServers": {
    "mauricekleine-nonobench": {
      "type": "http",
      "url": "https://www.nonobench.com/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.

### 2026-11-04 (score 86, 0)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-10-29 (score 86, +1)

- [security] Stability: 0.97 → pass

### 2026-10-27 (score 85, +1)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-10-25 (score 84, +1)

No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 83, 0)

- [security] The server rewrote its instructions, which are the text every model session reads

### 2026-10-23 (score 83, +1)

No change was recorded against any check on this day. Stability & Change Management went from 77 to 80. That category is still filling its 30-day observation window: 23 days of observed history at the previous scan, 24 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-21 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-19 (score 81, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (11)

### `get_leaderboard` (~185 tokens)

Get leaderboard

Models ranked by accuracy; defaults to all effort levels for compatibility.

Input parameters:

- `effort` (string): best, all (default), or one effort level; empty means all
- `family` (string): Comma-separated family ids; empty means no filter
- `min_correct` (integer): Minimum puzzles solved in the selected tier; default 0 includes unsolved variants
- `open_weights` (boolean): Only open-weight (true) or closed (false) models
- `provider` (string): Comma-separated provider ids; empty means no filter
- `reasoning` (boolean): Only reasoning (true) or non-reasoning (false) variants
- `size` (string): Grid size to filter on
- `version` (string): Comma-separated benchmark versions: 1.0, 1.1, 1.2; empty means all

Output parameters:

- `models` (array)
- `size` (string)
- `updatedAt` (string)

### `list_providers` (~19 tokens)

List providers

Provider ids, names, families and variant counts.

Output parameters:

- `providers` (array)

### `list_families` (~19 tokens)

List families

Model families, available efforts and best variants.

Output parameters:

- `families` (array)

### `compare_models` (~69 tokens)

Compare models

Side-by-side core overall and per-size accuracy, cost, latency and token results for model or family names.

Input parameters:

- `models` (array, required): 2 to 20 model variant ids or family names, e.g. claude-opus-5.5 or gpt-6-astra-xhigh

Output parameters:

- `models` (array)

### `get_model_results` (~53 tokens)

Get model results

Accuracy, cost, latency and token use for one model, broken down by grid size.

Input parameters:

- `model` (string, required): Model name as listed on the leaderboard, e.g. gpt-5.4-xhigh

Output parameters:

- `accuracy` (number): Percentage of puzzles solved
- `bySize` (array)
- `correct` (number)
- `displayName` (string)
- `effort` (string|null)
- `failedRuns` (number)
- `family` (string)
- `model` (string): Model variant id, e.g. claude-opus-5.5-high
- `provider` (string|null)
- `reasoning` (boolean)
- `total` (number)

### `list_puzzles` (~33 tokens)

List puzzles

The benchmark puzzles with their ids and row/column clues.

Input parameters:

- `size` (string): Grid size to filter on

Output parameters:

- `puzzles` (array)

### `get_puzzle` (~62 tokens)

Get puzzle

One puzzle, including the clue text models were prompted with. The reference solution is only included on request; some puzzles have several valid solutions.

Input parameters:

- `id` (string, required): Puzzle id from list_puzzles
- `include_solution` (boolean): Include a reference solution

Output parameters:

- `columnClues` (array)
- `height` (number)
- `id` (string)
- `index` (number)
- `prompt` (string): Clue text as models received it
- `referenceSolution` (string)
- `rowClues` (array)
- `size` (string)
- `url` (string)
- `width` (number)

### `check_solution` (~74 tokens)

Check solution

Check a grid against a puzzle's clues, using the same rule as the benchmark grader. Reports which rows and columns do not match.

Input parameters:

- `grid` (string, required): Row-major string of 0 (empty) and 1 (filled), width × height characters
- `id` (string, required): Puzzle id from list_puzzles

Output parameters:

- `columnViolations` (array)
- `correct` (boolean)
- `error` (string)
- `rowViolations` (array)

### `get_puzzle_results` (~142 tokens)

Get puzzle results

Per-model outcomes for one puzzle. Answers are omitted unless requested.

Input parameters:

- `effort` (string): best, all (default), or one effort level
- `family` (string): Comma-separated family ids; empty means no filter
- `id` (string, required): Puzzle id from list_puzzles
- `include_answers` (boolean): Include each model's answer grid
- `open_weights` (boolean): Only open-weight (true) or closed (false) models
- `provider` (string): Comma-separated provider ids; empty means no filter
- `reasoning` (boolean): Only reasoning (true) or non-reasoning (false) variants

Output parameters:

- `attempts` (number)
- `index` (number)
- `puzzleId` (string)
- `runs` (array)
- `size` (string)
- `solved` (number)
- `updatedAt` (string)

### `get_model_puzzles` (~55 tokens)

Get model puzzles

Which puzzles one model solved, missed, timed out on, or has not run.

Input parameters:

- `model` (string, required): Model variant id as listed on the leaderboard, e.g. claude-opus-5.5-high

Output parameters:

- `attempted` (number)
- `displayName` (string)
- `model` (string)
- `puzzles` (array)
- `solved` (number)

### `list_runs` (~114 tokens)

List runs

Individual benchmark runs (one model on one puzzle), optionally with the raw prompt and model output.

Input parameters:

- `include_output` (boolean): Include raw prompt and model output (large)
- `limit` (integer): Default 100
- `model` (string): Only runs of this model variant id
- `offset` (integer): Runs to skip, for paging; default 0
- `puzzle_id` (string): Only runs on this puzzle id from list_puzzles
- `size` (string): Grid size to filter on

Output parameters:

- `limit` (number)
- `offset` (number)
- `runs` (array)
- `total` (number)

## Diagnostics

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

## Score history

- 2026-11-05: 86
- 2026-11-04: 86
- 2026-11-03: 86
- 2026-11-02: 86
- 2026-11-01: 86
- 2026-10-31: 86
- 2026-10-30: 86
- 2026-10-29: 86
- 2026-10-28: 85
- 2026-10-27: 85
- 2026-10-26: 84
- 2026-10-25: 84
- 2026-10-24: 83
- 2026-10-23: 83
- 2026-10-22: 82
- 2026-10-21: 82
- 2026-10-20: 81
- 2026-10-19: 81
- 2026-10-18: 80
- 2026-10-17: 80
- 2026-10-16: 80
- 2026-10-15: 79
- 2026-10-14: 79
- 2026-10-13: 78
- 2026-10-12: 78
- 2026-10-10: 77
- 2026-10-04: 74
- 2026-10-03: 73
- 2026-10-02: 73
- 2026-10-01: 73

## Common questions

### What is the Nonobench MCP server?

Nonobench is an MCP server listed in the public MCP registry as io.github.mauricekleine/nonobench. An open-source benchmark of how well LLMs solve nonogram puzzles, from 5x5 to 20x20. This page covers its hosted endpoint (https://www.nonobench.com/mcp).

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

Nonobench scores 86 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 Nonobench MCP server expose?

Nonobench exposes 11 tools: get_leaderboard, list_providers, list_families, compare_models, get_model_results, and 6 more. Their descriptions and schemas cost roughly 825 tokens of context every time the server is loaded.

### Does the Nonobench MCP server require authentication?

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

### Is the Nonobench MCP server still maintained?

Nonobench 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://www.nonobench.com/mcp
- Repository: https://github.com/mauricekleine/nonobench
- Website: https://www.nonobench.com/
- Changelog RSS feed: https://verifymcp.io/servers/mauricekleine-nonobench/www.xml
- Changelog JSON feed: https://verifymcp.io/servers/mauricekleine-nonobench/www.json
- HTML version of this page: https://verifymcp.io/servers/mauricekleine-nonobench/www