Understanding Agents

Agent Definition Records

An Agent Definition is the core record that drives every AI agent in REDA AI. Think of it as the blueprint for your agent — it stores the name, purpose, capabilities, and configuration settings that determine how your agent behaves, where it can be deployed, and what it can access.

With AI included, select the agent’s Processing Tier. With Bring your own AI, select AI Connection and Model in the agent’s AI Account section, or inherit the organisation default. Different agents can use different supported providers and models. Prompts, tools, record context and channel permissions still define each agent’s job. Configure the agent’s AI account.

Key Agent Properties

Agent Configuration Elements

1. Persona & Prompts — Define the agent's identity, role, tone, and behavioral instructions through the Persona field and Custom Prompts.
2. Tools — Assign which tools the agent can use to query and modify Salesforce data.
3. Context Fields — Specify which Salesforce record fields automatically load into the agent's context on record pages.
4. Channels — Choose where the agent is deployed (Internal, External Widget, Community, or API).
5. Settings — Configure rate limits, message length, token budgets, timeouts, reasoning depth (Deep Reasoning), and other operational parameters.
6. Context Suggestions — Add quick-action buttons that users can click to send pre-written prompts.

Managing Multiple Agents

With AI included, select the agent’s Processing Tier. With Bring your own AI, select AI Connection and Model in the agent’s AI Account section, or inherit the organisation default. Different agents can use different supported providers and models. Prompts, tools, record context and channel permissions still define each agent’s job. Configure the agent’s AI account.

• Department-based agents — Separate agents for Sales, Service, Support, HR, Finance, etc.
• Channel-based deployment — One agent for internal Salesforce users (Internal channel) and another for external website visitors (External Widget channel).
• Persona variations — Agents with different tones and specializations (a friendly onboarding bot vs. a formal compliance assistant).
• Feature testing — Create test agents in sandbox to experiment with new prompts, tools, or settings before deploying to production.

The maximum number of active agents is determined by your REDA AI plan tier. Use the Active Status toggle to stay within your limit while keeping inactive agents available for future use.

Agents can consult each other

Splitting work across several agents raises an obvious question: what happens when a user asks the general assistant something only the specialist agent knows? An agent can consult another agent mid-conversation and use the answer, rather than guessing or telling the user to go somewhere else. The consulted agent answers using its own instructions, its own tools and its own data access — which is the point, because that expertise is exactly what it was built to hold.

This changes how you might divide agents up. Rather than one large agent that has to know everything, you can keep a focused assistant that knows who to ask. It stays under your control: an agent is only available to be consulted if you have explicitly made it so, and consultation only happens when the calling agent’s own instructions tell it to.

For how it works, the two settings that enable it, and what it costs, see Agent-to-Agent Delegation.

Agent Lifecycle

Cloning & Reusing Agents

Salesforce's standard clone functionality is available for Agent Definition records. Cloning is useful when you want to create a new agent based on an existing configuration — for example, cloning your Sales agent to create a similar agent for a different region or product line. After cloning, you can modify the persona, tools, and other settings to suit the new agent's purpose.

The Underlying AI Model

With AI included, select the agent’s Processing Tier. With Bring your own AI, select AI Connection and Model in the agent’s AI Account section, or inherit the organisation default. Different agents can use different supported providers and models. Prompts, tools, record context and channel permissions still define each agent’s job. Configure the agent’s AI account.

Common Agent Scenarios

Connect selected agents through MCP

MCP connects an approved AI assistant to selected active Internal agents. Each person signs in with their Salesforce identity. The host receives agent entry points; individual REDA tool and prompt definitions are not exposed as separate MCP tools. Configure a Salesforce Site, an External Client App, both REDA AI User and REDA AI MCP User assignments, and agent opt-in. See the administrator setup guide and user guide. Interaction data can reach the connected host and configured AI providers as well as Salesforce. Review the intended data and agreements; this is not a promise that data never leaves Salesforce.

Understand the Economy credit change

Economy ordinary usage now applies a 0.40 multiplier to the base credit rate. For unchanged token quantities and base rates, an example costing 100 base credits costs 40 Economy credits. At the previous 0.10 multiplier it cost 10: the new debit is four times the previous Economy debit, and 60% below base. This changes credit consumption, not the subscription price. File pre-extraction is charged separately at full rate. A provider fallback can use a different effective tier; inspect the actual transaction rather than applying 0.40 to every mixed operation. New credit transactions record Processing Tier; historical rows are not backfilled. Economy still requires consent and compatible licence/provider configuration; a missing consent is refused, not silently processed at another tier.

Frequently Asked Questions

Next Steps

Now that you understand the structure of agents, learn how to create your first agent or explore agent configuration options in detail.