Skip to content

DeepTeam is a framework to red team LLMs and AI agents.

.

penguine943eaac85e97
diff --git a/docs/docs/frameworks-mitre-atlas.mdx b/docs/docs/frameworks-mitre-atlas.mdx
index c2b2efa5..53af8733 100644
--- a/docs/docs/frameworks-mitre-atlas.mdx
+++ b/docs/docs/frameworks-mitre-atlas.mdx
@@ -324,29 +324,15 @@ red_team(
This stage ensures your system's final outputs and decisions remain aligned with **intended ethics, safety, and trust boundaries** — even under adversarial manipulation.
-## Assessing Your Production Application with MITRE ATLAS
-
-Red teaming your deployed applications is crucial to ensure your LLM applications are secure and compliant. You can run MITRE ATLAS-aligned assessments on your production application on Confident AI by directly connecting your application to the platform.
-
-[Confident AI](https://www.confident-ai.com) allows you to directly chat with your deployed application through a postman style connection. This allows you to run MITRE ATLAS assessments against your production application:
-
-<BlogImageDisplayer
- src={ASSETS.confidentSettingsProjectAiConnection}
- alt="AI Connection in Confident AI"
- caption="AI Connection in Confident AI"
-/>
-
-Confident AI also provides you with the full risk assessment report after running the assessment.
-
-### Risk assessment reports and PDF export
-
-After a run, Confident AI aggregates results into a **risk assessment** with CVSS-style severity, coverage, and per-test-case detail in the UI. You can also generate a **downloadable risk assessment report (PDF)** to share with security, compliance teams when you need a fixed artifact beyond the live dashboard.
+:::info Run MITRE ATLAS Assessments on Confident AI
+[Confident AI](https://www.confident-ai.com) lets you configure the MITRE ATLAS framework, schedule recurring risk assessments, manage vulnerabilities in one place, and share downloadable PDF reports with your team for security alignment.
<BlogImageDisplayer
src={ASSETS.confidentRedTeamingRiskAssessment}
alt="Risk assessment overview in Confident AI after a framework-based assessment"
caption="Example: consolidated risk assessment in Confident AI after running a framework-based assessment."
/>
+:::
## Best Practices
diff --git a/docs/docs/frameworks-nist-ai-rmf.mdx b/docs/docs/frameworks-nist-ai-rmf.mdx
index 71841327..3515f3ac 100644
--- a/docs/docs/frameworks-nist-ai-rmf.mdx
+++ b/docs/docs/frameworks-nist-ai-rmf.mdx
@@ -269,29 +269,15 @@ risk_assessment = red_team(
This category ensures AI systems not only measure risks effectively but also **close the feedback loop** by aligning transparency, accountability, and business value.
-## Assessing Your Production Application with NIST AI RMF
-
-Red teaming your deployed applications is crucial to ensure your LLM applications are secure and compliant. You can run NIST AI RMF-aligned assessments on your production application on Confident AI by directly connecting your application to the platform.
-
-[Confident AI](https://www.confident-ai.com) allows you to directly chat with your deployed application through a postman style connection. This allows you to run NIST AI RMF assessments against your production application:
-
-<BlogImageDisplayer
- src={ASSETS.confidentSettingsProjectAiConnection}
- alt="AI Connection in Confident AI"
- caption="AI Connection in Confident AI"
-/>
-
-Confident AI also provides you with the full risk assessment report after running the assessment.
-
-### Risk assessment reports and PDF export
-
-After a run, Confident AI aggregates results into a **risk assessment** with CVSS-style severity, coverage, and per-test-case detail in the UI. You can also generate a **downloadable risk assessment report (PDF)** to share with security, compliance teams when you need a fixed artifact beyond the live dashboard.
+:::info Run NIST AI RMF Assessments on Confident AI
+[Confident AI](https://www.confident-ai.com) lets you configure the NIST AI RMF framework, schedule recurring risk assessments, manage vulnerabilities in one place, and share downloadable PDF reports with your team for security alignment.
<BlogImageDisplayer
src={ASSETS.confidentRedTeamingRiskAssessment}
alt="Risk assessment overview in Confident AI after a framework-based assessment"
caption="Example: consolidated risk assessment in Confident AI after running a framework-based assessment."
/>
+:::
## Best Practices
diff --git a/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx b/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx
index 412a036e..eb9ff12c 100644
--- a/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx
+++ b/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx
@@ -411,29 +411,15 @@ critical_assessment = red_team(
)
```
-## Assessing Your Production Application with OWASP ASI
-
-Red teaming your deployed applications is crucial to ensure your LLM applications are secure and compliant. You can run OWASP Top 10 for Agentic Applications assessments on your production application on Confident AI by directly connecting your application to the platform.
-
-[Confident AI](https://www.confident-ai.com) allows you to directly chat with your deployed application through a postman style connection. This allows you to run OWASP ASI assessments against your production application:
-
-<BlogImageDisplayer
- src={ASSETS.confidentSettingsProjectAiConnection}
- alt="AI Connection in Confident AI"
- caption="AI Connection in Confident AI"
-/>
-
-Confident AI also provides you with the full risk assessment report after running the assessment.
-
-### Risk assessment reports and PDF export
-
-After a run, Confident AI aggregates results into a **risk assessment** with CVSS-style severity, coverage, and per-test-case detail in the UI. You can also generate a **downloadable risk assessment report (PDF)** to share with security, compliance teams when you need a fixed artifact beyond the live dashboard.
+:::info Run OWASP ASI Assessments on Confident AI
+[Confident AI](https://www.confident-ai.com) lets you configure the OWASP ASI framework, schedule recurring risk assessments, manage vulnerabilities in one place, and share downloadable PDF reports with your team for security alignment.
<BlogImageDisplayer
src={ASSETS.confidentRedTeamingRiskAssessment}
alt="Risk assessment overview in Confident AI after a framework-based assessment"
caption="Example: consolidated risk assessment in Confident AI after running a framework-based assessment."
/>
+:::
## Best Practices for Agentic AI Security
diff --git a/docs/docs/frameworks-owasp-top-10-for-llms.mdx b/docs/docs/frameworks-owasp-top-10-for-llms.mdx
index 201765a1..aee02823 100644
--- a/docs/docs/frameworks-owasp-top-10-for-llms.mdx
+++ b/docs/docs/frameworks-owasp-top-10-for-llms.mdx
@@ -393,29 +393,15 @@ print(f"Total test cases: {len(owasp_assessment.test_cases)}")
print(f"Pass rate: {owasp_assessment.pass_rate:.1%}")
```
-## Assessing Your Production Application with OWASP
-
-Red teaming your deployed applications is crucial to ensure your LLM applications are secure and compliant. You can run OWASP Top 10 assessments on your production application on Confident AI by directly connecting your application to the platform.
-
-[Confident AI](https://www.confident-ai.com) allows you to directly chat with your deployed application through a postman style connection. This allows you to run OWASP Top 10 assessments against your production application:
-
-<BlogImageDisplayer
- src={ASSETS.confidentSettingsProjectAiConnection}
- alt="AI Connection in Confident AI"
- caption="AI Connection in Confident AI"
-/>
-
-Confident AI also provides you with the full risk assessment report after running the assessment.
-
-### Risk assessment reports and PDF export
-
-After a run, Confident AI aggregates results into a **risk assessment** with CVSS-style severity, coverage, and per-test-case detail in the UI. You can also generate a **downloadable risk assessment report (PDF)** to share with security, compliance teams when you need a fixed artifact beyond the live dashboard.
+:::note Run OWASP Assessments on Confident AI
+[Confident AI](https://www.confident-ai.com) lets you configure the OWASP Top 10 framework, schedule recurring risk assessments, manage vulnerabilities in one place, and share downloadable PDF reports with your team for security alignment.
<BlogImageDisplayer
src={ASSETS.confidentRedTeamingRiskAssessment}
alt="Risk assessment overview in Confident AI after an OWASP-style framework run"
caption="Example: consolidated risk assessment in Confident AI after running a framework-based assessment."
/>
+:::
## Best Practices
diff --git a/docs/docs/red-teaming-introduction.mdx b/docs/docs/red-teaming-introduction.mdx
index 4f5a13bf..fe7647c7 100644
--- a/docs/docs/red-teaming-introduction.mdx
+++ b/docs/docs/red-teaming-introduction.mdx
@@ -673,24 +673,13 @@ The response evaluation process also involves two key stages:
The attacks are fed into the LLM, and the resulting responses are evaluated using vulnerability-specific metrics based on the types of attacks. **Each vulnerability has a dedicated metric** designed to assess whether that particular weakness has been effectively exploited, providing a precise evaluation of the LLM's performance in mitigating each specific risk.
-:::tip
It's worth noting that using a synthesizer model like GPT-3.5 can prove more effective than GPT-4o, as more **advanced models tend to have stricter filtering mechanisms**, which can limit the successful generation of adversarial attacks.
-:::
-
-`deepteam` runs locally in your development environment. [**Confident AI**](https://www.confident-ai.com) lets you run the same red teaming against production endpoints from the cloud — with scheduled runs, shareable risk reports, and no local infrastructure to maintain.
-## Red Teaming on the Cloud
-
-Connect your application once via an AI Connection (an HTTP endpoint that mirrors the `model_callback` contract), pick a framework like OWASP or MITRE ATLAS, and run assessments on demand or on a recurring schedule. Each run produces a risk report with CVSS-style severity scores and per-test-case analysis.
-
-<BlogImageDisplayer
- src={ASSETS.confidentRedTeamingChooseFramework}
- alt="Choosing a security framework in Confident AI"
-/>
-
-The comparison view across runs is where the value compounds — when you update a system prompt, swap models, or change retrieval logic, you can see exactly which vulnerabilities regressed and which improved. PDF exports are available for compliance documentation, security reviews, and stakeholder handoffs.
+:::note Red Team on Confident AI
+[Confident AI](https://www.confident-ai.com) lets you run red teaming on the cloud — configure frameworks like OWASP or MITRE ATLAS, schedule recurring risk assessments, manage vulnerabilities in one place, and share PDF reports with your team for security alignment.
<BlogImageDisplayer
src={ASSETS.confidentRedTeamingRiskAssessment}
alt="Risk assessment dashboard in Confident AI"
/>
+:::