io.github.barneywohl/bay-run MCP Server

io.github.barneywohl/bay-run provides a free OpenAI-compatible inference service with signed provenance receipts and three focused MCP tools. It is described as a decision and execution layer for durable specialist Pins at https://run.huggingbay.xyz, exposed via a canonical Streamable HTTP MCP endpoint.

🛠️ Key Features

Free OpenAI-compatible inference
Signed provenance receipts
Default MCP “front door” exposes three tools in order: coprocessor, run_pin, solve_task
Canonical endpoint: https://run.huggingbay.xyz/mcp/

🚀 Use Cases

Running durable specialist Pins from the canonical origin (https://run.huggingbay.xyz)
Using the MCP tools coprocessor, run_pin, and solve_task as the entry points

⚡ Developer Benefits

Streamable HTTP MCP endpoint for integration
Published request contract and auth instructions at https://run.huggingbay.xyz/quickstart
Signed provenance receipts for supported inference responses

⚠️ Limitations

Source content specifies toolCount as 2, while the default tool list contains three tools; the exact tool count is not consistently defined in the provided excerpt

{ "type": "object", "properties": { "task_description": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Plain-language task to solve; Bay Run infers the model family. Alias: task, task_hint. Support-ticket routing/triage requests with no caller-supplied candidate_labels or labeled examples reuse the existing provisional canonical Pin route_571826c40685073a99510b1951e60338; explicit labels/examples and sentiment, polarity, urgency, or unrelated classification remain on their requested path.", "title": "Task Description" }, "input": { "default": null, "description": "Task input. Send text/list for ordinary tasks or an object with query/documents, content, candidate_labels, or schema. Alias: query, text, content.", "title": "Input" }, "task": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Alias for task_description.", "title": "Task" }, "query": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "Alias for input or structured query.", "title": "Query" }, "text": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "Alias for input.", "title": "Text" }, "content": { "anyOf": [ {}, { "type": "null" } ], "default": null, "description": "Alias for input.", "title": "Content" }, "task_hint": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "default": null, "description": "Alias for task_description.", "title": "Task Hint" }, "examples": { "anyOf": [ { "items": { "additionalProperties": true, "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "Optional labeled input/expected examples.", "title": "Examples" }, "my_examples": { "anyOf": [ { "items": { "additionalProperties": true, "type": "object" }, "type": "array" }, { "type": "null" } ], "default": null, "description": "Alias for examples.", "title": "My Examples" } }, "title": "solve_taskArguments" }

{ "type": "object", "properties": { "pin_id": { "description": "Public strong Pin id. Canonical ids are listed in the tool description.", "maxLength": 128, "minLength": 3, "title": "Pin Id", "type": "string" }, "input": { "description": "Input matching the pinned task.", "title": "Input" }, "idempotency_key": { "anyOf": [ { "maxLength": 200, "minLength": 8, "type": "string" }, { "type": "null" } ], "default": null, "description": "Optional stable retry key. Omit and it is auto-minted from {pin_id, input}.", "title": "Idempotency Key" }, "max_price_usd": { "anyOf": [ { "maximum": 100, "minimum": 0, "type": "number" }, { "type": "null" } ], "default": null, "description": "Optional caller-authorized USD ceiling. Omit to inherit the pin's mint ceiling.", "title": "Max Price Usd" } }, "required": [ "pin_id", "input" ], "title": "run_pinArguments" }

Quickstart: connect to MCP

Use the canonical Streamable HTTP MCP endpoint: https://run.huggingbay.xyz/mcp/. The live copy-paste auth and request contract is kept at https://run.huggingbay.xyz/quickstart.

Mint a resource-bound demo bearer, then make the primary Guard-first call:

bash

BASE=https://run.huggingbay.xyz
TOKEN=$(curl -fsS -X POST "$BASE/oauth/token" \
  -H 'Content-Type: application/json' \
  -d '{"grant_type":"urn:bay-run:grant-type:demo","scope":"mcp:demo","resource":"https://run.huggingbay.xyz/mcp/"}' \
  | python3 -c 'import json,sys; print(json.load(sys.stdin)["access_token"])')

curl -fsS -X POST "$BASE/mcp/" \
  -H "Authorization: Bearer $TOKEN" \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"coprocessor","arguments":{"user_text":"<untrusted request>"}}}'

coprocessor runs the canonical Guard first and never generates text or executes tools. Add documents when retrieved chunks need bounded post-SAFE reranking. Read its action and evidence before deciding whether a caller-owned generation or tool call may continue.

Default MCP tools

Point an MCP client at https://run.huggingbay.xyz/mcp/. The public default surface is:

coprocessor — the primary bounded Guard-first call over user_text and optional documents.

run_pin — the direct alias for one of the four canonical Pins, using its published {pin_id,input} shape.

solve_task — the open-ended fallback when no canonical Pin matches; provide task_description and input.

Advanced compatibility capabilities remain available through explicit live references:

Hugging Bay mirror catalog

Hugging Bay’s mirror catalog is available at https://huggingbay.xyz. Use each artifact's published download plan for the currently admitted files. A drop-in HF_ENDPOINT compatibility origin is not part of the public contract yet, so this repository does not advertise one prematurely.

For current catalog and serving information, use the generated Bay Run status snapshot. Mirror counts and family breakdowns are intentionally not copied here because they change.

Install

Tools (2)

solve_taskrun_pin

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://run.huggingbay.xyz/mcp/

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "bay-run": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://run.huggingbay.xyz/mcp/"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

Bay Run

io.github.barneywohl/bay-run MCP Server

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

solve_task

run_pin