BrightOnAGENTIC Guide

Welcome to BrightOnAGENTIC Documentation

Everything you need to secure autonomous agents, and unlock continuous monitoring with BrightOnAGENTIC. Follow the guided steps, recommended workflows, and integration tips below.

Before you begin

Confirm you have BrightOnLABS admin access, a cloud provider service account, and at least one target agent workflow identified.

Account Setup

Create an organization in the BrightOnAGENTIC console and invite your security, platform, and compliance stakeholders. Assign at least one Owner and one Analyst to keep approvals and investigations separated.

Navigate to Settings → Organization to provision environments (Sandbox, Staging, Production).
Generate a service token under Settings → API Access for automated agent discovery.
Install the BrightOnAGENTIC Cloud Scanner in your preferred cloud account using the guided Terraform module.

                            curl -X POST
                            https://api.brightonagentic.com/v1/tokens \ -H
                            "Authorization: Bearer YOUR_ADMIN_TOKEN" \ -H
                            "Content-Type: application/json" \ -d '{ "label":
                            "sandbox-discovery", "scopes": ["agents.read",
                            "risks.write"] }'
                        

Store the returned token in your secrets manager; you will reference it when wiring BrightOnAGENTIC to CI/CD pipelines or external alerting.

Workspace Overview

The application is organized into four primary surfaces. Understanding them ensures faster triage and smoother collaboration.

Command Center

Real-time view of monitored agents, incident status, and posture drift. Use filters to switch between environments.

Agent Inventory

Every agent, model, and workflow BrightOnAGENTIC discovers, enriched with ownership, business criticality, and last scan time.

Risk Canvas

Graph visualization that maps dependencies, exposed data stores, and permission chains to surface blast radius.

Control Library

Pre-built guardrails, prompts, and verification checks you can apply to agents or bundle into automation playbooks.

Asset & Risk Mapping

BrightOnAGENTIC continuously correlates agent actions with the infrastructure and data they touch.

Connect cloud accounts and SaaS systems (Slack, Jira, GitHub) to ingest activity logs.
Tag agents with business impact scores to prioritize remediation plans.
Review suggested guardrails in the Risk Canvas and accept or customize before deployment.

Pro Tip

Use saved views to group agents by initiative (e.g., “Support Bots” or “Finance Automations”). This lets you assign ownership quickly when new risks appear.

Automations & Playbooks

Playbooks orchestrate actions when risks are detected or when scheduled hygiene checks run.

Build triggers from posture changes, anomalous prompts, or policy violations.
Chain actions such as notifying teams, rolling back configuration, or quarantining an agent.
Simulate playbooks in Sandbox mode before enabling in Production.

Guardrail Bundles

Start with the prebuilt “Tier 0 Agent Hardening” bundle. It enforces protected prompt segments, outbound domain allow-lists, and data redaction for sensitive entities.

Integrations

Extend BrightOnAGENTIC with existing observability, ticketing, and communication platforms.

SIEM & SOAR

Forward normalized alerts to Splunk, Microsoft Sentinel, or Chronicle using the streaming webhook collector.

Ticketing

Create incidents automatically in Jira Service Management or ServiceNow with deep links back to the affected agent.

ChatOps

Publish risk digests to Slack and Microsoft Teams channels with actionable buttons for acknowledge or escalate.

API Access

Use the REST API to export agent posture to your data warehouse on a schedule. See /v1/reports/posture.

Security & Compliance

BrightOnAGENTIC is designed for regulated environments. Key commitments include:

Data residency options in the US and EU with encrypted storage (AES-256 at rest, TLS 1.2+ in transit).
Role-based access control with SCIM user provisioning and SAML SSO.
Audit trails for every configuration change, retained for 18 months by default.
Independent SOC 2 Type II and ISO 27001 attestations renewed annually.

Contact security@brightonx.ai for penetration test reports or to request a dedicated tenant.

FAQ

How often are agents rescanned?

Discovery runs continuously; a full posture evaluation is executed every four hours by default. You can adjust cadence per environment.

Can BrightOnAGENTIC block unsafe actions in real time?

Yes. Enable inline enforcement in the Control Library to intercept prompts or outbound requests that violate guardrails.

What happens if an integration token expires?

The system raises a health alert under Command Center → Integrations and pauses automations tied to that connector until renewed.

Documentation coming soon

We’re preparing a full BrightOnAGENTIC knowledge base with onboarding steps, integration guides, and security playbooks. Check back shortly for the complete docs.