Claude Connector

Nonobench

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

Brand: Nonobench

Category: AI
Primary Subcategory: AI Project Directories & Agent Benchmarks

Integration details

Description

An open-source benchmark of how well LLMs solve nonogram puzzles, from 5x5 to 20x20. Query every result through a REST API or MCP server.

Integration type: Connector
Verification status: Community connector
Platform: Claude
Category: AI
Primary Subcategory: AI Project Directories & Agent Benchmarks
Secondary Subcategories: None listed
Brand: Nonobench
Access: No account required
First tracked: 2026-09-29
Tool count: 11
Geography: US

Agent DiscoverabilityClaude Discoverability Score

Claude Connector discovery is coming soon

Community Connectors only appear in Claude’s registry.Once verified, organic discovery begins and we can calculate your score.Read more about Community Connector verification.

No spam. Unsubscribe any time.

What discovery looks like

Claude showing a list of relevant Connectors in a conversation

Claude can surface verified Connectors when they match a user’s Prompt.

Primary Subcategory

Competing in Claude AI Project Directories & Agent Benchmarks

View Category

0 scoredVerified connectors by score · Community connectors A–Z

Primary Subcategory

—ArcopolisPending VerificationGet notified

—AttentionWatchPending VerificationGet notified

—Mixer4AIPending VerificationGet notified

—NonobenchPending VerificationGet notified

—PodPending VerificationGet notified

—PromptedSitePending VerificationGet notified

—StacklistPending VerificationGet notified

Tools

11 tools agents can invoke

Check a grid against a puzzle's clues, using the same rule as the benchmark grader. Reports which rows and columns do not match.
check_solution
Side-by-side core overall and per-size accuracy, cost, latency and token results for model or family names.
compare_models
Models ranked by accuracy; defaults to all effort levels for compatibility.
get_leaderboard
Which puzzles one model solved, missed, timed out on, or has not run.
get_model_puzzles
Accuracy, cost, latency and token use for one model, broken down by grid size.
get_model_results
One puzzle, including the clue text models were prompted with. The reference solution is only included on request; some puzzles have several valid solutions.
get_puzzle
Per-model outcomes for one puzzle. Answers are omitted unless requested.
get_puzzle_results
Model families, available efforts and best variants.
list_families
Provider ids, names, families and variant counts.
list_providers
The benchmark puzzles with their ids and row/column clues.
list_puzzles
Individual benchmark runs (one model on one puzzle), optionally with the raw prompt and model output.
list_runs

FAQs

Nonobench FAQ

How the directory, categories and Discoverability Score work.

Read the methodology

How do I improve a Community connector's discoverability?

The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.

What are Nonobench alternatives on Claude?

As of 2026-10-01, Nonobench competes with Arcopolis, AttentionWatch, Mixer4AI, Pod, PromptedSite, Stacklist in Claude AI Project Directories & Agent Benchmarks, ranked by public Discoverability Score.

Where is this profile measured?

This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.