OpsConduit Jobber MCP
Customer-hosted, read-only MCP server for Jobber owner/admin operational queries.
OpsConduit Jobber MCP
Read-only MCP connector for Jobber field-service operations.
OpsConduit helps owners and operators ask practical questions about the tools that run their business. This first connector is for Jobber and is designed to stay read-only by default.
Plain buyer-facing overview: docs/index.md. Common buyer questions are answered in docs/faq.md. Example owner/operator prompts are in docs/operator-prompts.md. First paid-install deliverable shape: docs/first-report.md.
What It Should Answer
- Which estimates need follow-up?
- Which invoices are overdue?
- Which jobs are unscheduled, stale, or blocked?
- Which customers have not had activity recently?
- What work needs attention this week?
- Where is operational follow-up leaking revenue?
More buyer-facing examples are in docs/use-cases.md.
A fictional sample output is available in docs/demo-output.md, or run:
npm run demo
The paid pilot handoff format is documented in docs/pilot-deliverable-template.md.
Product Posture
- Customer-hosted first.
- Read-only first.
- No credential custody required for the public repo.
- No autonomous writes in customer systems.
- No scraping private data.
- No fake claims, fake testimonials, or deceptive outreach.
Launch Offer
Early-access pricing is documented in docs/offer.md.
For pilot setup, email [email protected] with the subject Jobber pilot. Do not send Jobber passwords, tokens, API secrets, or private customer data by email.
Install
Requires Node.js 20+.
mkdir opsconduit-jobber
cd opsconduit-jobber
npm install @opsconduit/jobber-mcp
cp node_modules/@opsconduit/jobber-mcp/.env.example .env
npx opsconduit-jobber-auth
npx opsconduit-jobber-mcp
Fill JOBBER_CLIENT_ID and JOBBER_CLIENT_SECRET in .env before running the auth helper. The helper prints a Jobber OAuth URL, receives the local callback at http://localhost:3333/oauth/callback, and stores tokens in .opsconduit/jobber-tokens.json.
Detailed setup notes are in docs/install.md. Paid pilot validation steps are in docs/pilot-checklist.md. Implementation notes for Jobber's GraphQL API are in docs/jobber-api-notes.md.
MCP Tools
jobber_auth_url- create the one-time Jobber OAuth authorization link for a customer-hosted install.jobber_exchange_code- exchange the redirected OAuth code and store tokens locally without exposing them in chat.jobber_refresh_token- refresh the stored access token for continued read-only reporting.jobber_account- validate the connected account before running operational queries.jobber_clients_sample- validate read access with a capped client sample.jobber_graphql_read- run read-only Jobber GraphQL for owner reports; mutations are blocked before sending.
Policies
Status
This repository is in early implementation. The first usable target is a customer-hosted MCP server with local OAuth setup and read-only Jobber GraphQL access.
Servidores relacionados
Kone.vc
patrocinadorMonetize your AI agent with contextual product recommendations
JiraMCP
Full Jira & Confluence CRUD
Israeli Bank MCP
Manage Israeli bank accounts and transactions.
Google Calendar
Integrate Google Calendar with enhanced security using OAuth2 credentials.
Spotify MCP Server
Control Spotify with natural language. Enables search, playback control, queue management, and device control using conversational commands.
Notion
Interact with Notion's API to read, create, and modify content using natural language.
PDF2MD.node (mcp-node.io)
Privacy-first PDF to Markdown/JSON converter for AI Agents. Optimized for complex B2B invoices (E-Fatura) and tables. Features native Turkish character correction and local-first WASM parsing.
YouTube Video Summarizer
Fetch and summarize YouTube videos by extracting their titles, descriptions, and transcripts.
Date-time Tools
A server for date-time manipulation and timezone conversion.
omniparser-autogui-mcp
An MCP server that analyzes the screen with OmniParser to automate GUI operations.
PaperlessMCP
MCP server for Paperless-ngx document management. 43 tools for AI-powered document organization - full CRUD on documents, tags, correspondents, document types, storage paths, and custom fields.