Connect JobNimbus to Claude: Track Conversations & Work Activities

If your team needs to connect JobNimbus to Claude to automate SMS messaging, track field work activities, or audit project financials, you need a Model Context Protocol (MCP) server. This server acts as the translation layer between Claude's tool-calling capabilities and the JobNimbus REST API. You can either build and maintain this infrastructure yourself, or use a managed integration platform like Truto to dynamically generate a secure, authenticated MCP server URL. If your team uses ChatGPT, check out our guide on connecting JobNimbus to ChatGPT or explore our broader architectural overview on connecting JobNimbus to AI Agents.

Giving a Large Language Model (LLM) read and write access to a sprawling CRM and project management ecosystem like JobNimbus is an engineering challenge. You have to handle authentication lifecycles, map massive JSON schemas to MCP tool definitions, and navigate JobNimbus-specific domain constraints. Every time the upstream API changes, you have to update your server code, redeploy, and test the integration.

This guide breaks down exactly how to use Truto to generate a secure, managed MCP server for JobNimbus, connect it natively to Claude Desktop, and execute complex workflows using natural language.

The Engineering Reality of the JobNimbus API

A custom MCP server is a self-hosted integration layer. While the open MCP standard provides a predictable way for models to discover tools, the reality of implementing it against specialized B2B APIs is painful. JobNimbus is built to manage complex contracting businesses, field workforces, and multi-channel communications. Its API reflects that complexity.

If you decide to build a custom JobNimbus MCP server, here are the specific integration challenges you will face:

The Conversation and Message Duality JobNimbus does not simply have a "send text" endpoint. Communications are strictly hierarchical. To send an SMS to a customer, you must first resolve or create a Conversation record linking a contact's phone number to one of your agent phone numbers (agentPhoneNumber). Once the conversation is established and linked to a primary record (like a Contact or Job), you must then execute a distinct API call to create a Message inside that conversation. An LLM cannot infer this state machine. Your MCP server must expose strictly defined tools that enforce this sequence.

Strict Primary Record Constraints on Activities The JobNimbus API strictly enforces relational integrity on its activity feeds. If you attempt to list activities, the API outright rejects requests that fail to filter by either a primaryRecordId or createdById. You cannot simply ask the API for "all recent activities." Furthermore, certain activities generated by system integrations or marked with read-only activity types cannot be modified or deleted. Your MCP tools must provide clear JSON Schema constraints and descriptions that instruct the LLM on exactly which parameters are required to prevent continuous 400 Bad Request errors.

JSON Patch for Bulk Operations JobNimbus heavily utilizes JSON Patch (application/json-patch+json) for bulk updates and partial record modifications. For example, updating multiple conversations or patching a credit memo requires constructing a specific array of operations (e.g., [{"op": "replace", "path": "/assignedUserId", "value": "12345"}]). LLMs are notoriously bad at unprompted JSON Patch construction. Your MCP tool definitions must explicitly define the allowed paths and expected payload structures to prevent hallucinated payloads.

Transparent Rate Limiting Accountability When integrating AI agents, rate limits are a critical failure point. A single LLM loop can burn through an API quota in seconds. It is important to note how Truto handles this: Truto does not retry, throttle, or apply backoff on rate limit errors. When the JobNimbus API returns an HTTP 429, Truto passes that error directly to the caller. Truto normalizes upstream rate limit info into standardized headers (ratelimit-limit, ratelimit-remaining, ratelimit-reset) per the IETF spec. The caller (your agent framework) is strictly responsible for implementing its own retry and backoff logic. Do not expect the infrastructure to magically absorb these limits.

Generating the JobNimbus MCP Server

Truto derives MCP tools dynamically from documentation and API schemas. Rather than hand-coding tool definitions for JobNimbus, Truto exposes tools based on verified integration resources. A tool only appears in the MCP server if it has a corresponding documentation entry - this acts as a quality gate ensuring the LLM receives accurate JSON Schemas.

You can generate an MCP server for JobNimbus using either the Truto UI or the API.

Method 1: Via the Truto UI

For ad-hoc agent configuration or internal testing, generating a server through the dashboard is the fastest route.

1. Log into your Truto account and navigate to your connected JobNimbus Integrated Account.
2. Click the MCP Servers tab.
3. Click Create MCP Server.
4. Select your desired configuration (e.g., allow read and write methods, set tag filters if necessary).
5. Click Save and copy the generated MCP server URL. This URL contains a secure, hashed token.

Method 2: Via the Truto API

For programmatic, multi-tenant deployment, you can generate MCP servers dynamically for your users. This is ideal when embedding AI features into your own SaaS platform.

Make a POST request to /integrated-account/:id/mcp with your desired configuration:

curl -X POST https://api.truto.one/integrated-account/{integrated_account_id}/mcp \
  -H "Authorization: Bearer YOUR_TRUTO_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "JobNimbus Claude Server",
    "config": {
      "methods": ["read", "write", "custom"]
    }
  }'

The API returns a payload containing the secure endpoint URL:

{
  "id": "mcp_token_abc123",
  "name": "JobNimbus Claude Server",
  "config": {
    "methods": ["read", "write", "custom"]
  },
  "expires_at": null,
  "url": "https://api.truto.one/mcp/xyz789securetoken..."
}

Connecting the MCP Server to Claude

Once you have your Truto MCP URL, you need to provide it to Claude. The process differs slightly depending on how you use Claude.

Option A: Via the Claude UI (Desktop/Web)

If you are using Claude Desktop or Claude Web for personal or team use:

1. Open Claude and navigate to Settings -> Integrations (or Connectors depending on your tier).
2. Click Add MCP Server or Add custom connector.
3. Paste the Truto MCP URL (https://api.truto.one/mcp/...) you generated earlier.
4. Click Add. Claude will automatically perform the JSON-RPC handshake and ingest the available JobNimbus tools.

Option B: Via Manual Config File (Claude Desktop / Agent Frameworks)

If you are configuring Claude Desktop locally for development, you can add the server directly to your claude_desktop_config.json file. Because Truto serves the MCP over Server-Sent Events (SSE), you use the official @modelcontextprotocol/server-sse package.

{
  "mcpServers": {
    "jobnimbus_truto": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-sse",
        "https://api.truto.one/mcp/xyz789securetoken..."
      ]
    }
  }
}

Restart Claude Desktop. The JobNimbus tools are now natively available in your prompts.

JobNimbus MCP Hero Tools

Truto normalizes the JobNimbus API into granular, LLM-friendly tools. Here are the highest-leverage tools available for orchestrating field workflows and communications.

list_all_job_nimbus_conversations

Retrieves a list of SMS conversations. This is heavily filterable by assigned user, archived state, linked contact, and phone numbers. Crucial for agent context gathering before sending messages.

"Find all unread SMS conversations assigned to John Smith regarding recent roofing inquiries."

create_a_job_nimbus_conversation

Finds or creates an SMS conversation between a contact phone number and one of your agent phone numbers. This is a prerequisite step before a message can be dispatched.

"Create a new text conversation between our main support line and phone number 555-0199, linking it to the primary contact record for Sarah Connor."

create_a_job_nimbus_message

Sends a message inside a specific conversation. Supports scheduling for later delivery and attaching media.

"Send a text message in conversation ID 84920 letting the customer know our field technician is 15 minutes away."

list_all_job_nimbus_activities

Fetches the activity feed (notes, task updates, status changes) attached to a specific primary record or created by a specific user. The API strictly requires either a primaryRecordId or createdById.

"Pull the complete activity history for Job ID 49281 to see if the site inspection notes have been logged yet."

create_a_job_nimbus_activity

Logs a new activity attached to a primary record. Essential for ensuring the LLM leaves a trail of its automated actions in the CRM for human oversight.

"Log a new activity note on Contact ID 11223 stating that the automated follow-up SMS sequence was successfully initiated."

create_a_job_nimbus_credit_memo

Generates a financial credit memo against a primary record. Requires idempotency keys to prevent duplicate financial records.

"Create a credit memo for $150 against Job ID 77334 to account for the delayed material delivery, using 'delay_comp_001' as the idempotency key."

job_nimbus_conversations_bulk_update

Executes bulk updates on several conversations simultaneously using a JSON Patch document. Useful for sweeping administrative cleanups.

"Take these five conversation IDs and bulk update them to be archived and unassigned, since the project is now closed."

For the complete inventory of available JobNimbus tools - including query schemas, body schemas, and required fields - visit the JobNimbus integration page.

Workflows in Action

MCP tools are powerful in isolation, but they become transformative when chained together by an LLM reasoning engine. Here is how Claude executes real-world JobNimbus workflows.

Scenario 1: Automated Post-Site Visit Follow-Up

Field technicians log completion notes in JobNimbus, but administrative staff often forget to follow up with customers. You can prompt Claude to handle the outbound communication and log the interaction.

"Check the recent activities for Job ID 99283. If the site inspection was marked complete today, find the SMS conversation for the associated contact and send them a text thanking them for their time and letting them know the estimate will be ready tomorrow. Finally, log a note on the job that you sent this message."

How Claude executes this:

1. Calls list_all_job_nimbus_activities passing the Job ID to verify the inspection status.
2. Calls list_all_job_nimbus_conversations filtering by the Job's linked contact ID to find the active SMS thread.
3. Calls create_a_job_nimbus_message to dispatch the outbound text.
4. Calls create_a_job_nimbus_activity to log an audit trail note on the Job record.

sequenceDiagram
    participant User as Claude Desktop
    participant MCP as Truto MCP Server
    participant API as JobNimbus API
    
    User->>MCP: Call tool: list_all_job_nimbus_activities
    MCP->>API: GET /activities?primaryRecordId=99283
    API-->>MCP: Return activity history
    MCP-->>User: Inspection complete status
    
    User->>MCP: Call tool: list_all_job_nimbus_conversations
    MCP->>API: GET /conversations?contactId=123
    API-->>MCP: Return conversation ID
    MCP-->>User: ID: 88291
    
    User->>MCP: Call tool: create_a_job_nimbus_message
    MCP->>API: POST /messages
    API-->>MCP: 201 Created
    MCP-->>User: Success
    
    User->>MCP: Call tool: create_a_job_nimbus_activity
    MCP->>API: POST /activities
    API-->>MCP: 201 Created
    MCP-->>User: Audit note logged

Scenario 2: Bulk Auditing and Archiving Stale Communications

Support queues in JobNimbus can get cluttered with old, abandoned SMS threads. An admin needs to clean up the workspace.

"Find all text conversations assigned to agent ID 4455 that haven't had a message in the last 30 days, and bulk archive them to clear out the inbox."

How Claude executes this:

1. Calls list_all_job_nimbus_conversations using query parameters for the agent ID and a date filter for lastMessageAt.
2. Parses the returned array of conversations to extract the IDs.
3. Constructs a JSON Patch array mapping each ID to an archive operation.
4. Calls job_nimbus_conversations_bulk_update to execute the sweep in a single network request.

Security and Access Control

Giving an LLM unconstrained access to a CRM and financial system like JobNimbus is a massive security risk. Truto's MCP architecture provides native guardrails enforced at the server level, preventing the LLM from hallucinating destructive actions.

• Method Filtering (config.methods): Restrict servers to specific operations. A server configured with methods: ["read"] will only expose list and get tools, physically preventing the LLM from mutating data.
• Tag Filtering (config.tags): Scope access to specific functional areas. For example, filtering by ["communication"] ensures the LLM can manage conversations and messages but cannot touch credit memos or core CRM records.
• Required Authentication (require_api_token_auth): By default, possessing the MCP URL grants access. By enabling this flag, the client must also pass a valid Truto API token via the Authorization header, adding a strict identity check.
• Ephemeral Access (expires_at): Generate time-bound MCP servers for contractors or temporary AI agents. Once the ISO datetime is reached, the server self-destructs, severing access immediately.

Architecting for Scale

Building an AI agent that speaks to JobNimbus requires more than just formatting API requests. You have to handle OAuth tokens, manage the nuance of JobNimbus's nested conversation models, and ensure your system respects strict JSON schemas for activities and bulk operations.

Truto handles the authentication, schema mapping, and tool generation layers so your engineering team can focus on agent reasoning and prompt design. By deploying a managed MCP server, you instantly grant Claude secure, documented access to the entire JobNimbus ecosystem with zero custom integration code.