Identity Control Plane: The Unifying Layer for Zero Trust Infrastructure

Surya Teja Avirneni · 2025

Read Identity Control Plane: The Unifying Layer for Zero Trust Infrastructure by Surya Teja Avirneni (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-17759.txt.

Introduction

Surya Teja Avirneni wrote this security paper in 2025, a theory piece on zero trust infrastructure. Zero trust systems grant no access by default, and every request must prove itself. Avirneni proposes an Identity Control Plane, a single layer that speaks for all comers: people, workloads, and automation. It blends three ID kinds: workload IDs from SPIFFE, user IDs from web standards, and scoped robot credentials.

Brokers issue short transaction tokens under rules set by policy engines such as OPA and Cedar. The paper lays out the parts, join patterns, and use cases, then weighs the design against current models. It also maps the plan onto FedRAMP and SLSA compliance needs. No build is reported; the aim is a blueprint for researchers and platform architects.

Catalog record

CREATORSurya Teja Avirneni
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE18 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-17759

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Identity Control Plane: The Unifying Layer for Zero Trust Infrastructure Surya Teja Avirneni Abstract Identity has emerged as the foundational control vector in modern Zero Trust infrastructure. However, identity is often fragmented across human, workload, and automation domains, leading to inconsistent policy enforcement, static privileges, and limited auditability. This paper introduces the Identity Control Plane (ICP), a unifying architectural pattern that integrates SPIFFE-based workload identity, OIDC/SAML user identity, and scoped automation tokens. ICP enables dynamic, intent-aware access control using attribute-based policy engines. We describe its core components, modes of enforcement, use cases, and integration patterns. We also include a comparative analysis with existing models and theorize performance characteristics such as token evaluation latency and enforcement complexity. The ICP aligns with IETF WIMSE standards and provides a scalable framework for secure, identity-aware infrastructure orchestration. I Introduction Zero Trust architectures prioritize identity-aware access control across distributed, multi-tenant systems. As organizations adopt internal developer platforms (IDPs), CI/CD pipelines, and multi-cloud infrastructure, the surface area of identity increases significantly. Today, identity is fragmented: human users authenticate through OIDC or SAML, workloads may use SPIFFE or IAM roles, and automation often relies on static secrets or long-lived service accounts. This results in inconsistent policy enforcement, over-permissioning, and poor auditability. This paper introduces the Identity Control Plane (ICP) —a unifying control layer that abstracts identity across domains and enforces fine-grained, dynamic policies based on actor attributes, runtime metadata, and organizational intent. The ICP model integrates: • SPIFFE-based workload identity for service-to-service authentication [ 1 ] , • OIDC/SAML-based identity for human actors, • Broker-issued transaction tokens for scoped automation credentials, • Attribute-based access control (ABAC) enforcement engines such as OPA [ 3 ] and Cedar [ 7 ] . This work builds on concepts from SPIFFE, the OAuth transaction token model, and the IETF WIMSE initiative, which standardizes cross-system workload identity [ 4 , 5 , 6 ] . We present the ICP architecture, use cases, and integration points with cloud-native systems. While we do not provide a full prototype, we include a comparative analysis with existing solutions and theorize performance properties including token validation latency, enforcement scope, and policy drift risks. II Background and Related Work II-A Workload Identity and SPIFFE The Secure Production Identity Framework for Everyone (SPIFFE) standard [ 1 ] enables workload-to-workload authentication using X.509 or JWT-based service identity documents (SVIDs). SPIRE, its reference implementation, provides workload attestation and trust domain management. SPIFFE decouples identity from infrastructure topology and supports service mesh enforcement (e.g., Istio), but does not unify identity across humans and automation. II-B Human Identity: OIDC and SAML OpenID Connect (OIDC) and SAML are widely adopted for user authentication and federated access. While effective for SSO and role assignment, these models lack ephemeral, intent-bound identity constructs and are unsuitable for runtime authorization or scoped policy enforcement in automation and infrastructure workflows. II-C IETF WIMSE: Workload Identity in Multi-System Environments The IETF WIMSE working group defines standards for federated identity and credential exchange across systems [ 4 ] . Its Workload Identity Token and Credential Exchange drafts [ 5 , 6 ] provide a foundation for secure, context-rich authentication between actors operating in distinct trust domains. The Identity Control Plane adopts and operationalizes these principles. II-D ABAC and Policy Engines Attribute-based access control (ABAC) has gained adoption through engines like Open Policy Agent (OPA) [ 3 ] and Cedar [ 7 ] . These systems evaluate policies using actor metadata, environment context, and resource tags. However, they require a consistent identity substrate and integration with runtime execution layers to enforce policies reliably. II-E Credential Brokers and Automation Identity Credential brokers issue scoped, short-lived tokens for automation tasks, typically through external identity verification and context evaluation. Prior work on Zero Trust CI/CD [ 8 ] and credential delegation [ 9 ] highlighted the need for ephemeral, intent-scoped automation identity in pipelines and ephemeral environments. II-F Related Architectures and Gaps TABLE I: Comparison of Identity Enforcement Models Model Scope Federation Tokenization Policy Enforcement AWS IAM Workload/User Limited Static RBAC Cloud API Istio + SPIFFE Workload Yes SVID RBAC Proxy GitHub OIDC User/Automation Medium OIDC JWT Limited CI/CD ICP (This Work) All Full Scoped ABAC Multi-Layer As shown in Table I, existing systems either focus on one actor type or lack runtime enforcement. The Identity Control Plane fills this gap by offering a composable framework for real-time, policy-driven identity orchestration across diverse control planes. III Problem Definition Cloud-native infrastructure operates across multiple trust domains, identity sources, and operational planes. Despite widespread adoption of Zero Trust principles, identity systems remain disjointed and inconsistent in their enforcement capabilities. III-A Fragmented Identity Models Human, workload, and automation identities are issued and verified by different systems—OIDC for users, SPIFFE or cloud IAM for services, and shared secrets for automation. These silos inhibit secure delegation and increase operational complexity. III-B Lack of Intent-Bound Credentials Automation workflows often rely on long-lived service accounts or static role bindings. These credentials do not encode transaction context (e.g., commit SHA, actor, purpose), making fine-grained enforcement and post-hoc attribution difficult. III-C Static Policy Models RBAC and IAM policies are typically static and environment-agnostic. They lack support for runtime metadata such as deployment state, environment tags, or triggering pipeline. This leads to over-permissioning and brittle policy definitions. III-D No Unified Control Layer There is no central mechanism to normalize identity across systems and enforce attribute-based policies at the point of execution. Existing IDPs provide authentication but not identity-bound enforcement or simulation at the infrastructure layer. III-E Audit and Compliance Blind Spots Without a shared identity substrate, access logs are fragmented. Most systems cannot map resource access back to the initiating identity in a verifiable, cryptographically secure way. This inhibits auditability and regulatory compliance. Summary: A scalable Zero Trust model requires a unified, identity-aware control plane that enables scoped, intent-driven authorization and enforcement across users, workloads, and automation. IV Identity Control Plane Architecture The Identity Control Plane (ICP) unifies authentication, tokenization, and policy enforcement for all identity types—human, workload, and automation. It abstracts identities across trust domains and evaluates access dynamically using attribute-based policies. IV-A Core Components 1) Identity Abstraction Layer: Normalizes identity formats using: • SPIFFE IDs for workloads [ 1 ] • OIDC/SAML for human users • Broker-issued transaction tokens for automation and CI/CD jobs 2) Credential Brokers: Generate scoped, short-lived tokens. Tokens embed metadata including actor, source system, environment, time, and purpose. 3) Policy Evaluation Layer: Enforces access using ABAC engines like OPA [ 3 ] and Cedar [ 7 ] . Policies reference identity, resource metadata, and real-time intent (e.g., environment, git ref, action type). 4) Metadata Integration: Pulls context from internal developer platforms (IDPs), source control, Kubernetes metadata, and cloud control planes. IV-B Modes of Operation • Real-Time Enforcement: Policy evaluated at the moment of action (e.g., API request, deployment, shell access). • Intent Evaluation: Dry-run simulation to validate access before execution (e.g., PR checks, deployment preview). • Replay and Audit: Post-event simulation to explain policy behavior or validate policy changes retroactively. IV-C SPIFFE Bundle Synchronization for Cross-Domain Trust To enable secure cross-domain workload authentication, SPIFFE uses trust bundles that define the root CA certificates for other trust domains [ 1 ] . The Identity Control Plane incorporates bundle synchronization mechanisms—either

Read the complete text