Salesforce Service Cloud
Salesforce Service Cloud is a customer service platform providing case management, knowledge base, omnichannel support, and service automation to deliver exceptional customer experiences
Verdict
Common use cases
- Triage high-priority cases during standup
- Draft customer replies with full case history
- Update ticket statuses from chat threads
- Pull SLA breach reports for weekly reviews
- Log call notes directly into case records
Integration
- Vendor
- Salesforce Service Cloud
- Category
- crm
- Auth
- OAUTH2
- Composio slug
salesforce_service_cloud
Tools
Setup
Setup guide
- 11. In Switchy, navigate to Settings > Integrations and select Salesforce Service Cloud from the CRM category. 2. Click Connect and you'll be redirected to Salesforce's OAuth consent screen. 3. Sign in with your Service Cloud credentials and grant read/write access to Cases, Contacts, and Accounts (scopes: api, refresh_token). 4. After authorization, Switchy redirects you back and confirms the connection is active. 5. Open any Space and type '@Salesforce Service Cloud list open cases' to test — you should see a response pulling live case data. 6. To invoke tools in conversation, @mention the MCP by name followed by your request (e.g. '@Salesforce Service Cloud update case 00123456 status to In Progress'). 7. If the connection fails, verify your Salesforce admin has enabled API access for your user profile.
What teammates see: by default, memories from Salesforce Service Cloud are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
List Urgent Cases
@Salesforce Service Cloud show me all open cases with priority set to High or Critical, sorted by created date descending.Open in a Space →
Summarize Customer History
@Salesforce Service Cloud pull the last five cases and any open tasks for contact email john.doe@example.com.Open in a Space →
Update Case Status
@Salesforce Service Cloud update case number 00123456 to status 'Waiting on Customer' and add a comment that we sent follow-up questions.Open in a Space →
Check SLA Compliance
@Salesforce Service Cloud list all cases where the SLA violation date is within the next 24 hours or already passed.Open in a Space →
Create New Case
@Salesforce Service Cloud create a new case for contact jane.smith@example.com with subject 'Billing discrepancy on invoice 9876' and priority Medium.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Salesforce Service Cloud show me all open cases assigned to the support team that haven't been updated in the last 48 hours
Found 7 open cases matching your criteria. Case #00123456 (Priority: High) from Acme Corp regarding API timeout errors, last updated 3 days ago. Case #00123457 (Priority: Medium) from Beta Industries about login issues, last updated 2 days ago. Case #00123458 (Priority: Low) from Gamma LLC requesting feature documentation, last updated 4 days ago. The remaining 4 cases are all marked Medium priority and range from 2-5 days since last activity. All cases are currently in 'New' or 'Working' status.
This example demonstrates querying case data with time-based filters. Requires read access to Case objects in your Salesforce org. The actual field names and case structure depend on your org's configuration and custom fields. Response time varies with case volume.
@Salesforce Service Cloud create a new case for customer account Delta Systems with subject 'Billing discrepancy on invoice INV-2024-0891' and assign it to the billing queue
Created Case #00123459 successfully. Subject: 'Billing discrepancy on invoice INV-2024-0891'. Account: Delta Systems (Account ID: 001xx000003DGb2AAG). Status: New. Priority: Medium (default). Assigned to: Billing Queue. Case number has been generated and the account contact will receive an automated email notification if your org's email-to-case settings are enabled. You can view the full case details in your Salesforce instance.
This write operation requires create permissions on Case objects and knowledge of your org's queue names. The MCP will use default field values unless you specify priority, case origin, or other attributes. Side-effect: triggers any active automation rules (assignment rules, workflows, email alerts) configured in your org.
@Salesforce Service Cloud summarize the resolution patterns for all closed cases tagged 'integration-error' in the last quarter and suggest what documentation we should update
Analyzed 23 closed cases tagged 'integration-error' from the past 90 days. Resolution pattern: 17 cases (74%) resolved by updating API credentials or refreshing OAuth tokens, average resolution time 4.2 hours. 4 cases (17%) required endpoint URL corrections in customer configurations, average resolution time 2.1 hours. 2 cases (9%) needed our engineering team to patch rate-limit handling. Recommendation: Update your API authentication guide to emphasize token expiration monitoring and add a troubleshooting section on common endpoint misconfiguration symptoms. Consider adding a pre-flight connection test tool to your integration setup wizard.
This synthesis example combines case data retrieval with AI reasoning to identify trends. Accuracy depends on consistent case tagging practices in your org. The AI interprets resolution notes and comments, so results improve when your team writes structured case updates. Requires read access to closed cases and their related objects.
Use-case deep-dives
When your support team needs AI to read case history before escalating
A 6-person support team fields 40 tickets a day. When a case needs engineering, the handoff context is scattered across Service Cloud notes, email threads, and Slack. This MCP lets Switchy pull the full case timeline—customer contact history, prior interactions, attached logs—so the AI can draft a handoff summary engineers actually read. The win is speed: your L1 agent types "summarize case 00123 for eng handoff" and gets a 4-sentence brief in 10 seconds instead of hunting through tabs. The threshold: if your team closes fewer than 20 cases a week, the OAuth setup overhead outweighs the time saved. For teams running triage at scale, this MCP turns case context into a one-prompt lookup.
When you need to spot churn signals buried in support volume
A 12-person CS team manages 80 accounts. Three customers opened 9 cases each in the last two weeks—normal ticket volume or a red flag? This MCP lets Switchy query Service Cloud for case frequency, sentiment tags, and resolution time by account, then surface accounts crossing your churn threshold. The AI can answer "which accounts opened more than 5 P1 cases this month" without a BI tool or custom report. The trade-off: if your team already has a dashboard for this, the MCP adds latency (OAuth + API calls take 2-4 seconds per query). Use this when your health-score logic changes faster than your BI team can rebuild reports, or when you need ad-hoc pattern checks mid-call.
When new hires need to search resolved cases without learning SOQL
A SaaS company hires 3 support reps every quarter. Training used to mean a week of shadowing plus a 40-page runbook. With this MCP, new reps ask Switchy "how did we handle login issues for enterprise customers last month" and get summaries of closed cases, including resolution steps and macros used. The MCP searches Service Cloud's case history and knowledge base without the rep learning Salesforce's query syntax or navigating 6 tabs. The boundary: if your knowledge base is under 50 articles, a static FAQ doc is faster to search. For teams with 200+ resolved cases a month and high rep turnover, this MCP turns case history into a searchable onboarding resource that updates itself.
Frequently asked
What does the Salesforce Service Cloud MCP do in Switchy?
It connects your Salesforce Service Cloud instance to Switchy so AI assistants can read case data, customer records, and service metrics without switching tabs. Your team can ask questions about open tickets, escalation queues, or SLA status and get answers pulled directly from Service Cloud. The MCP uses OAuth2, so you authenticate once and Switchy maintains the session.
Do I need Salesforce admin permissions to connect this MCP?
You need a Salesforce user account with API access enabled. Most Service Cloud licenses include API access, but your admin may have restricted it. If you can't see the OAuth consent screen when connecting, ask your Salesforce admin to grant API Enabled permission to your user profile. You don't need to be a full admin yourself.
Can the MCP create or update cases in Service Cloud?
Not yet. The current integration is read-only, so AI assistants can surface case details, customer history, and queue stats but can't modify records or create new cases. If your workflow requires writing back to Service Cloud, use Salesforce's native API or Zapier until Switchy adds write tools to this MCP.
How is this different from just logging into Salesforce directly?
You skip the context-switching. Instead of opening Service Cloud, running reports, and copying data into a chat, your team asks Switchy and the AI pulls the answer from Service Cloud in real time. It's faster for ad-hoc questions and keeps the conversation in one place, but you still use the Salesforce UI for bulk edits or complex workflows.
Who on the team should connect the Salesforce Service Cloud MCP?
Anyone who needs to query case data or customer records. Support leads and account managers get the most value because they field questions about ticket status and customer history all day. Each person connects their own Salesforce account, so permissions follow your existing Service Cloud roles. It doesn't consume extra Salesforce licenses.