Model Radar MCP server

Read from the package source without running it. The installed server may list more.

• askwrite action

  Run the same prompt on multiple models in parallel and return all responses.

• backtranslate_eval

  Evaluate a translation via back-translation and gloss overlap.

• batch_judgewrite action

  Run judge evaluations at scale on a list of items.

• batch_runwrite action

  Run multiple prompts in parallel with bounded concurrency and auto-retry.

• benchmark

  Quality-test models with 5 coding challenges and return pass/fail scores.

• compare

  Blind A/B comparison of two items judged by N models.

• configure_key

  Add or update a named API key. Does not delete other keys on this provider.

• credits

  Read leftover credits or quota where a public API exists.

• get_fastest

  Get the N fastest available models right now. Use when the user wants recommendations or "best/fastest/free" models.

• get_workers

  Get N verified-alive models across N distinct providers, ranked by tier then latency.

• host_swap_instructions

  Tell the host agent where to search the machine to swap in a model-radar model.

• judge

  Rate a single item using N diverse judge models and return aggregate scores.

• list_models

  List models in the catalog without pinging. Use when the user asks what models are available or to browse by tier/provider/free.

• list_providers

  List all providers with status (configured/unconfigured, kind, model count).

• profile

  Show spend lanes and local funded/spend_ok flags. No secrets.

• provider_status

  Check health of all configured providers by pinging one model from each.

• quality_probe

  Time and score a few models on a fixed probe for a job.

• recommend

  Recommend a short diverse lineup for a job (no ping).

• refresh_models

  Fetch latest model lists and replace each provider's catalog (add new ids, purge retired ones).

• restart_server

  Request the server to exit so a process manager can restart it (SSE only).

• runwrite action

  Run a prompt on the fastest available model and return the response.

• scan

  Ping models in parallel and return ranked results by latency. Use when you need live speed data or a ranked list.

• server_stats

  Return when this server process started and how long it has been running.

• set_profile

  Set local spend_ok / funded / login for a provider. Saved to ~/.model-radar/config.json.

• setup_guide

  Get setup instructions for adding free model providers.

• setup_workflow

  Deterministic setup workflow: guide the host to get API keys installed.

• still_free

  Lane A sweep: identify which default-pool hosts still answer.

• Code scan31 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitylocal package, no credentials required12/15
• Maintenancelast push 64 days ago12/15
• Maintainer identityregistry namespace matches repository owner7/10

Safety score

86/100

higher than 74% of 769 SEO servers

Tools

27

the typical SEO server has 6

Tools that change things

15%

typical: 0%. Fewer write tools means less an agent can break

Downloads a week

30

more than 19% of 170 SEO servers with a package

Model Radar: common questions

Is Model Radar MCP server safe?

Yes, by our scan: it is graded A (86/100). Read the Model Radar safety report

How do I install Model Radar?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Model Radar need an API key?

Not as far as the registry entry and our scan can tell: no credentials are declared or required.

Is Model Radar maintained?

The last commit was 65 days ago (2026-08-31). The latest release is v0.12.0.

What can I use instead of Model Radar?

Servers from other publishers that do the same job: DispatchSEO MCP server, Snapdec MCP server and DeepThonk MCP server.

How does Model Radar compare with other SEO MCP servers?

safety score 86/100 (higher than 74% of 769 SEO servers); tools 27 (the typical SEO server has 6); tools that change things 15% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.