dwaar use cases By agent

Sales agents

Let sales agents act on leads, not your entire CRM.

Sales agents research accounts, read emails, update CRM records, schedule meetings, and draft outreach. These connections create a path from untrusted external content to valuable customer and pipeline data.

By agent Sales agents Selected use case
dForge
Test before launch

Tests malicious lead data, injected instructions in emails and webpages, unauthorized discounts, bulk exports, and unintended outreach.

dFence
Enforce at runtime

Governs CRM, email, calendar, enrichment, and sales MCP calls according to the assigned account and requested task.

Intent + context One security model across the agent lifecycle
Sales agents in practice

Test the failure paths. Enforce the intended path.

Sales agents research accounts, read emails, update CRM records, schedule meetings, and draft outreach. These connections create a path from untrusted external content to valuable customer and pipeline data.

dForge / Before deployment

Find how the workflow can be manipulated.

Tests malicious lead data, injected instructions in emails and webpages, unauthorized discounts, bulk exports, and unintended outreach.

dFence / At runtime

Keep live actions aligned with intent.

Governs CRM, email, calendar, enrichment, and sales MCP calls according to the assigned account and requested task.

Example

A security decision in context.

A salesperson asks the agent to research Acme and draft a follow-up. A compromised webpage instructs it to export every CRM contact to an external webhook. dForge exposes the attack chain, while dFence blocks the export as an intent mismatch.

One security model

The use case changes. The security loop stays the same.

For every agent, dwaar asks four questions: What is the agent supposed to do? Which data and tools should it reach? How could that workflow be manipulated? What should happen when its behavior drifts?

01

Define

Document the happy path, permitted tools, data boundaries, and approval points.

02

Test

dForge attacks the workflow before launch using context-specific prompt, tool-use, and multi-turn tests.

03

Enforce

dFence evaluates runtime intent before sensitive tool and MCP calls, then allows, blocks, constrains, redacts, or requests approval.

04

Improve

Turn verified findings and runtime events into stronger policies and repeatable security tests.

Build with control

Know where your agent can be misused before it goes live.

dwaar combines agent-specific adversarial testing with intent-aware runtime enforcement, helping teams deploy capable agents without relying on static scopes alone.