Governance guide

How to govern AWS Bedrock

A step-by-step guide to governing AWS Bedrock with Rencore: detect with 30 policies, review with 15 reports, and remediate with 1 automations.

Definition

Governing AWS Bedrock means keeping its access, configuration, and lifecycle under continuous control rather than reacting after something breaks. Rencore governs AWS Bedrock with 30 pre-built policies, 15 reports, and 1 automations, so teams can detect risk, review posture, and remediate with an audit trail. The steps below turn that coverage into a repeatable routine.

Steps

  1. Inventory AWS Bedrock

    Connect AWS Bedrock and let Rencore build a continuous inventory of its resources, owners, and configuration, so governance starts from what exists rather than a stale export.

  2. Detect with policies

    Turn on the pre-built policies that cover AWS Bedrock to surface oversharing, sprawl, and misconfiguration on the first scan, before writing a single custom rule.

  3. Review by owner and severity

    Use the AWS Bedrock reports to review findings by owner, category, and severity, and to share them with stakeholders who do not have a seat in the platform.

  4. Remediate and automate

    Apply automations to fix findings at scale, route sensitive changes through approvals, and keep every action reversible and logged for the audit trail.

Recommended AWS Bedrock policies

Grounded in the Rencore catalog. See the full AWS Bedrock catalog on the AWS Bedrock connector page.

  • Agent is live and callable

    Agents in PREPARED state are callable, raising the likelihood that any weakness is exploited

    Severity: Medium
  • Flow is live and invocable

    Flows in Prepared state are invocable, raising the likelihood that any weakness is exploited

    Severity: Medium
  • Agent can perform actions

    Agents with an enabled action group can invoke external APIs, raising the likelihood of real-world impact

    Severity: High
  • IAM user has console access

    Users with interactive console access are a more readily exploited entry point

    Severity: Medium
  • IAM user has active access keys

    Users with active programmatic credentials are a more likely source of a credential leak

    Severity: Medium
  • Role assumable by Bedrock

    Roles assumable by the Bedrock service have live, runtime-exercised permissions

    Severity: Medium
  • AWS Bedrock Agent without Guardrail

    Detects agents that do not have a guardrail configured

    Severity: Medium
  • AWS Bedrock Agent using deprecated model

    Detects agents using foundation models with LEGACY or EOL lifecycle status

    Severity: High
  • AWS Bedrock Agent in failed state

    Detects agents with FAILED or NOT_PREPARED status

    Severity: High
  • AWS Bedrock Flow in failed state

    Detects flows with Failed status

    Severity: High
  • AWS Bedrock Knowledge Base in failed state

    Detects knowledge bases with FAILED status

    Severity: Medium
  • AWS Bedrock Agent not updated in 90 days

    Detects agents that have not been updated in the last 90 days

    Severity: Low
Explore the full AWS Bedrock governance catalog | All guides

Trusted by

MAPALBAMVille de LuxembourgWACKERGRUNDFOSAMGENOsramLufthansaThyssenKruppSunrisePattern