Fine Grained Authorisation with Relationship-Based Access Control
summary
Relationship-based access control is shown end-to-end: an EAV model stores entities, relationships and permissions; a local demo runs in Docker with Postman and a React UI against Permi; toggling a friendship flips a permission that lets a dog eat a bone. The design follows Google’s Zanzibar idea, yields automatic offboarding via relationship breaks, and returns snap tokens for cacheable state; note a Per mi limitation — it won’t traverse more than two relationship hops, so longer chains require aliasing.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 3:00Access control basics
- 4:24Role-based access control
- 5:29Attribute-based access control
- 8:25Relationship-based access control
- 12:01Under the hood of ReBAC
- 16:30Mapping ReBAC models
- 25:09Implementation and policy
- 30:04Real-world use cases
- 32:42Demo and technical setup
- 41:53Conclusion and considerations
- 47:38Q&A and wrap up
related talks
47mFine Grained Authorisation with Relationship Based Access Control
41mFine Grained Authorisation with Relationship-Based Access Control
54mFine-Grained Authorization for Modern Applications
30mFine-Grained Authorization: The Missing Piece in Agentic AI Security
33mHow to Prevent AI Agents from Accessing Unauthorized Data
55mYour AI Agents Are Overprivileged: The Case for Fine-Grained Authorization
44mSecuring RAG Pipelines with Fine Grained Authorization
37mRole-Based Access Is so Yesterday: Revolutionizing Authorization with Open FGA