Skip to main content
Glama
a-mart

Boneyard Integration Starter

by a-mart
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Boneyard integration starter

Build an MCP server that plugs into Boneyard as a connection, without changing Boneyard. This repository contains:

  • The contract: docs/contract.md. Language-agnostic rules for authentication, caller context, tool naming and annotations, errors and limits.

  • A reference server in TypeScript (Node 22, official MCP SDK, stateless Streamable HTTP). It has two example tools, tickets_search (read) and ticket_add_comment (write), backed by an in-repository fake.

  • The kit: kit check, a conformance suite you can run against any server URL, and kit dev, a mock Boneyard gateway for trying tools as a given agent, person and channel.

Coding agents: start with AGENTS.md. To build your own server, follow Building a new server and the Replace the example checklist, which lists every file that mentions the ticket example. The kit and its tests don't depend on the example, so they keep passing when you delete it.

Quick start

npm ci
npm run verify      # typecheck, standards, tests, and a full conformance run against the reference server
npm run dev         # serve http://127.0.0.1:8750/mcp with the fake backend
npm run kit:dev     # in a second terminal: call tools as Boneyard would

Check any server, including one in another language:

npm run kit -- check https://your-server.example.com/mcp --token-file ./connection-token

Related MCP server: TicketDesk MCP

Documentation

Page

For

contract.md

What your server must do (any language)

concepts.md

Connections vs integrations, Off, Ask and Allow, what the admin sees

backends.md

Backends without an API key (file shares, mounted folders), fixture roots, canaries, admin configuration

security.md

Secrets, least privilege, containment, prompt injection, audit, regulated data

testing.md

npm test, kit check (safe and fake modes), kit dev

deploying.md

Same host (Docker network), elsewhere over HTTPS, Windows service, file shares

registering.md

Adding the connection in Boneyard

server-json.md

Describing your headers so Boneyard renders a form

Configuration (reference server)

Variable

Default

Meaning

CONNECTION_TOKEN_FILE / CONNECTION_TOKEN

required

Token Boneyard sends as Authorization: Bearer ... (at least 32 characters)

BACKEND_API_KEY_FILE / BACKEND_API_KEY

required

Credential for the backend (the fake checks it too)

BACKEND

fake

Backend implementation

TRUST_CALLER_CONTEXT

false

Use Boneyard-* headers for attribution and logs

HOST / PORT / MCP_PATH

127.0.0.1 / 8750 / /mcp

Listener

MAX_RESULT_BYTES

65536

Result bound per tool call (at most 262144)

MAX_BODY_BYTES

1048576

Request body cap

ALLOWED_ORIGINS

empty

Browser origins allowed (the gateway sends none)

LOG_LEVEL

info

debug, info, warn or error

License

MIT. See LICENSE.

This server cannot be deployed

Pay for results, not tokens.Fund a bounty. Let AI agents compete and collaborate to solve your problems.agentbounties.appAd

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • TicketDesk MCP

    bhargavlukka
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables AI assistants to search and retrieve support tickets, add notes, close tickets, and access SLA policies through MCP tools, resources, and prompts over Streamable HTTP.
    MIT

  • Ticket Queue MCP Server

    sanjay1110
    F
    license
    Not graded
    quality
    C
    maintenance
    Enables AI clients to access and manage an internal support ticket queue through MCP tools, resources, and prompts, including searching and viewing tickets, adding comments, closing tickets with confirmation, reading knowledge base articles, and viewing queue summaries over OAuth-secured Streamable HTTP.
    -

  • ToolBridge

    Sehrabdar
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables autonomous AI agents to securely access and execute external tools, such as GitHub REST API operations, with per-user authentication, authorization, audit logging, and observability.
    MIT