Grabset

by abidlabsco.huggingface.abidlabs/grabhub

Grab samples from any HF dataset into a new set

  • Remote endpoint
  • 5 tools

Uptime

Checked  · every day

100%from 4 days of checks

Up without downtime on 4 of 4 days with data.

  • No downtime
  • Degraded
  • Down over 5 minutes
  • No data
  • Not checked yet

Response time

Median 1115 ms, slowest 5% 1259 ms for the handshake and the tool list, over 30 days.

1500 ms750 ms0

How far each check got

  • Reached the address 100%
  • Completed the handshake 100%
  • Listed its tools 100%

5 tools at the last answer.

Every day
DayResultChecksMedian
3 Nov (today)No downtime11186 ms
2 NovNo downtime1of 11048 ms
1 NovNo downtime1of 11114 ms
31 OctNo downtime21111 ms

A failure counts as down time only when a second check five minutes later fails too. How we check

Connect to the publisher's own endpoint

Add to ~/.continue/config.yaml:

mcpServers:
  - name: "grabhub"
    type: sse
    url: "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
For agents: the install prompt

Every step above as one Markdown text, for an assistant installing it for you.

# Install the MCP server Grabset (`co.huggingface.abidlabs/grabhub`)

Its page: https://plug.directory/servers/co.huggingface.abidlabs/grabhub

It runs on the publisher's servers; the client connects to its URL.

Use the section for the person's client.

## Claude Desktop

Paste into Settings › Connectors › Add custom connector:

```text
https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse
```

## Claude Code

Run in a terminal:

```sh
claude mcp add --transport sse grabhub https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse
```

## Cursor, Devin Desktop, LM Studio, Gemini CLI, oterm, ECA

Merge into the place for the person's client:

- Cursor: ~/.cursor/mcp.json
- Devin Desktop: ~/.config/devin/mcp_config.json
- LM Studio: mcp.json (Program › Install › Edit mcp.json)
- Gemini CLI: ~/.gemini/settings.json
- oterm: config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows)
- ECA: ~/.config/eca/config.json (global) or .eca/config.json in the project

```json
{
  "mcpServers": {
    "grabhub": {
      "url": "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## VS Code

Merge into .vscode/mcp.json:

```json
{
  "servers": {
    "grabhub": {
      "type": "sse",
      "url": "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Zed

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Cline, AnythingLLM, Augment Code

Merge into the place for the person's client:

- Cline: cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers)
- AnythingLLM: anythingllm_mcp_servers.json in the AnythingLLM storage plugins directory
- Augment Code: Augment Settings Panel – MCP section – Import from JSON (VS Code extension)

```json
{
  "mcpServers": {
    "grabhub": {
      "type": "sse",
      "url": "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Continue

Merge into ~/.continue/config.yaml:

```yaml
mcpServers:
  - name: "grabhub"
    type: sse
    url: "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
```

## Goose

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Codex

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## mcp-client-cli, TUUI, BoltAI, Bob Shell, Kiro, Kiro CLI

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Agenvoy

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Amp

Merge into ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):

```json
{
  "amp.mcpServers": {
    "grabhub": {
      "url": "https://abidlabs-grabhub.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

Then restart the client, or reload its MCP servers, and ask it to list the server's tools to confirm it started.

Tools 5

  • grabhub_build_image_set Build a LoRA-ready image set: relevant to the query, watermarks and near-duplicates removed, and spread out for variety (DINOv3 subject diversity).
  • grabhub_describe_audio_dataset Return the fingerprint profile of one catalog dataset: card, facet distributions (language, bandwidth, noise, duration, sounds) and summary stats over its sampled clips.
  • grabhub_route_datasets Rank whole Hugging Face datasets for a query, using their fingerprinted sample index. Returns, for each dataset, its repo, the config/split pairs that contain matching samples, and the best-matching indexed rows ("seeds"). The grabset Python client uses this to decide which rows to stream and embed on the fly.
  • grabhub_search_audio_samples Search individual samples across Hugging Face datasets by what they contain. Audio is the first modality indexed (10 datasets); other modalities return no results yet. Each sample was fingerprinted with an audio-LLM caption, CLAP and Jina omni audio embeddings, AudioSet sound tags, Whisper language ID and signal probes (bandwidth, SNR). Results mix samples from different datasets and include pointers (repo, config, split, row_idx) that load the audio with the dataset viewer or `datasets`.
  • grabhub_search_image_samples Search individual images across Hugging Face image datasets (styles, photography, characters, edit pairs) by what they show. Each result has a dense caption, structured attributes (medium, shot, camera angle, lighting, palette), aesthetic/NSFW/watermark scores, tags, and a pointer (repo, config, split, row_idx) to the source row.

From Hugging Face Spaces. Report a problem