AI & Machine Learning

huggingbay.xyz

Provides open metadata for AI artifacts including provenance, licenses, and inventory.

3 endpoints20 known toolsFirst detected July 25, 2026Last detected September 6, 2026

ENDPOINT 1

https://huggingbay.xyz/api/mcp

No auth detected

Known tools 0

No tool metadata was available in the registry cache.

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.huggingbay-xyz]
url = "https://huggingbay.xyz/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "huggingbay-xyz": {
      "type": "http",
      "url": "https://huggingbay.xyz/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: huggingbay-xyz
Remote MCP URL: https://huggingbay.xyz/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "huggingbay-xyz": {
      "url": "https://huggingbay.xyz/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "huggingbay-xyz": {
      "type": "http",
      "url": "https://huggingbay.xyz/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "huggingbay-xyz",
  "transport": "streamable-http",
  "url": "https://huggingbay.xyz/api/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

ENDPOINT 2

https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp

No auth detected

MCP server metadata

Name: bay-run
Version: 1.12.4

Capabilities

experimentalpromptsresourcestools

Known tools 15

discover_models

Search a 147K-model catalog for a small, cheap, open specialist model that does ONE narrow job better/cheaper than a general LLM — embeddings, reranking, text classification, NER/extraction, routing, guardrails, transcription, vision, audio.

Inferred read-only

eval_models

Prove which candidate model actually wins on YOUR data before committing — a head-to-head bake-off, not a public leaderboard (MTEB rank does NOT predict your-domain performance).

Inferred read-only

embed

Turn text into embedding vectors using ANY open embedding model, served instantly on demand — no packaging, no deploy, no GPU.

Inferred read-only

rerank

Reorder candidate documents by true relevance to a query using an open cross-encoder/reranker, served instantly on demand — the standard move to sharpen RAG / search precision after a vector search returns a noisy top-k.

Inferred read-only

extract

Turn messy HTML/text (e.

Inferred read-only

find_specialist_for_task

ONE call to find the best small specialist model for your task, proven on YOUR examples.

Inferred read-only

route

RUNTIME auto-router: send a job WITHOUT knowing which specialist to use, and Bay Run picks the best mirrored specialist per-request at inference time.

Potential side effects

classify

Classify text with a small CPU-served text-classification specialist — the guardrail / safety / moderation / sentiment / intent / NLI layer agents need and that frontier routers don't offer as tiny models.

Inferred read-only

request_specialist

Ask Bay Run for a specialist for a task — and NEVER get a dead end.

Inferred read-only

remember

DURABLE cross-call MEMORY: store a small key->value so future runs of YOUR agent keep context between calls (agents' #1 repeated pain — no built-in memory between tool calls).

Inferred read-only

recall

Read back DURABLE MEMORY you stored with `remember`.

Inferred read-only

forget

Delete DURABLE MEMORY.

Potential side effects

calculate

Evaluate arithmetic/math EXACTLY and instantly — the reliable calculator agents need because LLMs silently botch multi-step arithmetic.

Inferred read-only

validate_json

Validate JSON instantly: if `data` is a STRING, confirm it PARSES; if a `schema` (JSON Schema) is given, validate the value against it and return every violation.

Inferred read-only

resolve_link

Check whether a URL is ALIVE (fast HEAD/GET); if it's dead/unreachable AND it names a model Hugging Bay has MIRRORED, return the mirrored copy's serve pointer — a fallback UNIQUE to Bay Run (we hold the mirror).

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.bay-run]
url = "https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "bay-run": {
      "type": "http",
      "url": "https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: bay-run
Remote MCP URL: https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "bay-run": {
      "url": "https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "bay-run": {
      "type": "http",
      "url": "https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "bay-run",
  "transport": "streamable-http",
  "url": "https://bay-run-mvp-zfmlsu2yla-uc.a.run.app/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

ENDPOINT 3

https://run.huggingbay.xyz/mcp

No auth detected

MCP server metadata

Name: bay-run
Version: 2.0.0

Capabilities

experimentalpromptsresourcestools

Server instructions

Lead with solve_task when the caller has one task and input: one call routes, executes, and returns a result plus receipt. Use run_bakeoff to prove your model when the caller has 3-10 labeled examples: optionally put any public Hugging Face challenger beside the curated incumbents; one call infers the task family, evaluates them, and returns a signed winner plus a shareable public result. For recurring execution use get_task_quote, then run_task, then verify_result. Bay Run receives a specialized task and hard constraints, quotes the best qualifying route for free, executes without exceeding max_price_usd, and returns a verifiable result. Follow the response.next_call binding after each primary tool; raw JSON-RPC bindings use result.structuredContent. Mint the demo bearer with scope mcp:demo and resource https://run.huggingbay.xyz/mcp/. try_bay_run is a deprecated compatibility proof. Advanced compatibility tools remain directly callable and are documented at /.well-known/mcp/advanced-tools.json.

Known tools 5

solve_task

Route and execute one task-shaped call without learning quote/run shapes.

Potential side effects

run_bakeoff

ONE-CALL BAKE-OFF: give Bay Run 3-10 string/string labeled examples and no model knowledge.

Inferred read-only

get_task_quote

Get a FREE, expiring decision for a specialized task.

Inferred read-only

run_task

Execute an existing task quote, or quote-and-run from a complete task spec.

Potential side effects

verify_result

Verify a Bay Run execution and receipt without re-running the task.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.bay-run]
url = "https://run.huggingbay.xyz/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "bay-run": {
      "type": "http",
      "url": "https://run.huggingbay.xyz/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: bay-run
Remote MCP URL: https://run.huggingbay.xyz/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "bay-run": {
      "url": "https://run.huggingbay.xyz/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "bay-run": {
      "type": "http",
      "url": "https://run.huggingbay.xyz/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "bay-run",
  "transport": "streamable-http",
  "url": "https://run.huggingbay.xyz/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for huggingbay.xyz was fetched 2026-08-23T20:28:18.377Z.

Trust status Neutral

huggingbay.xyz is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.