liveBook · Manning

Cedar-agent

As explained in Authorization in Action

Cedar Local Agent is a reference implementation that exposes the Cedar engine through a REST API and manages policies and entities in memory.

A Cedar Local Agent (CLA) is an open-source Rust component that keeps Cedar policies and entities in a customizable local cache. A policy enforcement point (PEP) can call it over HTTP before forwarding protected operations to application logic. section 9.8

This arrangement suits services that need low-latency, fine-grained authorization without depending on a public cloud. The operational tradeoff is local responsibility for deployment, scaling, updates, and policy or entity distribution. section 9.8 section 11.1.5

A local agent is useful when:

  • A performance-sensitive service needs authorization without an additional remote network hop. section 11.1.5
  • An organization needs control over Cedar policy and entity lifecycles across environments. section 9.8
  • A service combines baseline network restrictions with resource-level Cedar decisions. section 11.1.5

Translating a document-sharing request

A PEP can turn a share request into an evaluation containing an employee principal, document-share action, document resource, and context such as managed-device status and request time. Cedar returns the authorization decision, which the PEP enforces before the backend runs. section 9.8

Figure 9.2 The Policy Enforcement Point (PEP) in ACME’s Customer Collaboration platform translates a real-world /share request into a Cedar evaluation by combining session data (principal), entity data (resource relationships), and context signals (environment attributes). Cedar’s is_authorized API provides an allow or deny decision, which the PEP enforces before forwarding the request to the backend.

Trace how session data, entity relationships, and environmental context become a Cedar decision enforced at the application boundary. section 9.8

How Authorization in Action covers it

How deep it goes
Explains deployment benefits, operational costs, and request integration, but defers detailed policy and entity distribution design.
Examples
Uses an ACME document-sharing evaluation with principal, action, resource, device, and time attributes.
What it assumes
Assumes familiarity with Cedar requests, policies, entities, and policy enforcement points while showing how evaluation data is assembled.
What you can do after
Operate/configure a local Cedar evaluation path and weigh its latency and lifecycle costs against a managed API.
How it connects
Connects Cedar to policy enforcement points, policies, entities, HTTP APIs, in-process enforcement, and Amazon Verified Permissions.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Authorization in Action. The section links point to the parts of the book it draws on.