Salesforce Agentforce: practical guide, pricing, alternatives and use cases
Machine-translated from Italian. Read the original
Salesforce AgentforceWhat is Salesforce Agentforce
Agentforce is Salesforce's platform for creating AI agents embedded directly in the CRM. The agents can respond to customers, support sellers, and automate tasks by leveraging data in Sales Cloud, Service Cloud, and Field Service.
Getting started
- Salesforce organization – you need to have a Salesforce organization. The Salesforce Foundations plan provides Agentforce Builder and Prompt Builder for free.
- Enable Agentforce – activate the Agentforce User License (paid) or use Flex Credits for initial testing.
- Define a use case – choose a narrow process (e.g., answering order FAQs) and configure the agent with Prompt Builder.
- Test and rollout – verify responses in a sandbox, set escalation rules to a human operator, and monitor cost per conversation.
- First useful result – for example, the agent automatically generates order confirmation emails with customer data pulled from the CRM.
Concrete use cases
- Customer support – automatic answers to questions about order status, shipping, and appointments, reducing load on the service desk.
- Sales assistant – generation of personalized sales emails with the latest contact information and opportunities from Sales Cloud.
- Field Service – suggestion of maintenance tasks based on asset data and service history.
- Lead qualification – the agent interrogates the prospect, gathers information, and creates or updates the lead in the CRM.
- Rapid reporting – automatic summarization of meetings or calls with data extracted from the CRM, ready to be shared.
Pricing and plans
| Plan | Price | Period | Notes |
|---|---|---|---|
| Salesforce Foundations | $0 | - | Agentforce Builder and Prompt Builder |
| Flex Credits | $500 | 100,000 credits | Pay‑per‑action (20 credits per action) |
| Conversations | $2 | per conversation | Agents toward customers |
| Agentforce User License | $5 | user/month | Requires Flex Credits |
| Agentforce add‑ons | $125 | user/month | Unlimited use for Sales, Service, Field Service |
Source: official pricing page, verified by Penzai on 13/11/2026. Prices may change: always check before purchasing.
Advantages and limitations
Advantages
- Native integration with all Salesforce modules (Sales, Service, Field Service).
- Consumption‑based model (credits or conversations) that allows gradual scaling.
- Visual tools (Agentforce Builder, Prompt Builder) with no coding required.
Limitations
- Requires an existing Salesforce organization; for non‑Salesforce users the entry cost is high.
- Consumption costs can grow quickly if thresholds or human escalations are not set.
- Advanced customizations (e.g., complex logic) require Apex skills or external integrations.
Alternatives
Other AI‑for‑CRM solutions exist; see the dedicated section for an updated overview.
- Docusign IAM by Docusign: electronic signature and agreement management with AI contract analysis.
- Zendesk AI by Zendesk: customer support software with AI agents that answer tickets.
- Apollo.io by Apollo: commercial prospecting platform with contact data and AI.
- Intercom Fin by Intercom: customer support with Fin, the AI agent that resolves requests.
- Glean by Glean: AI search and assistant on corporate data, respecting permissions.
Integrations and API
- Sales Cloud – access to opportunities, contacts, and accounts.
- Service Cloud – case management, knowledge base, and tickets.
- Field Service – asset data, service scheduling.
- Salesforce REST API – to call custom actions from the agent.
For additional integrations or API details, refer to the official Salesforce Agentforce documentation.
GitHub, open source and skills
There is currently no official GitHub repository for Agentforce.
- SalesforceCommerceCloud/b2c-developer-tooling: B2C CLI, VSCode Extension, Skills+MCP, and Tooling SDK for Salesforce Agentforce Commerce B2C (58 stars, TypeScript, Apache‑2.0 license)
- lucianostraga/sfagent-tools: AI‑powered testing for Salesforce Agentforce agents. Claude Code plugin. (24 stars, Apex)
- misu007/agentforce-headless-agent-nextjs-demo: A Next.js demo app showcasing integration with Salesforce Agentforce through the Agent API (9 stars, TypeScript)
- Beingwizard/Salesforce-Agentforce-Workshop: Build your retail agent with Salesforce Agentforce (8 stars)
- mvrzan/salesforce-agentforce-a2a-wrapper: Example implementation of a financial AI Agent built with Agentforce and exposed via the A2A protocol. The Heroku backend provides A2A protocol s… (7 stars, TypeScript)
GitHub data as of 14/11/2026. "Official" only if the repository belongs to the vendor's organization.
Skills to develop
- Basic: using Agentforce Builder and Prompt Builder.
- Intermediate: creating custom actions, managing credits, and configuring escalations.
- Advanced: integrating with Apex, building complex workflows, and monitoring organization‑wide costs.
Tips and deeper insights
- Start with a small FAQ set and measure average response time before expanding scope.
- Set daily credit limits and threshold notifications to avoid billing surprises.
- Use Prompt Versioning to track changes and test iterations.
- Involve sales and support teams in prompt definition: the language must reflect their way of working.
Community questions
- Which use case have you implemented with Agentforce and what metrics did you monitor?
- How did you manage consumption cost control in your organization?
- Have you integrated Agentforce with custom Apex processes? Share examples.