OpenFGA
FGA · Open FGA
An open-source fine-grained authorization (FGA) engine, modelled on Google's Zanzibar, that stores relationship tuples and answers yes/no access checks.
In our briefings
The claim each of the newest briefings on OpenFGA would have you leave with.
Drus sorts microservice authorization into five patterns, from his rework of OWASP's microservice security cheat sheet; the names are his own, and he says three were already in the sheet and two are new. Keeping the decision in a service's own code is fine, he says, when policy and data are local to that service. An external decision point such as Open Policy Agent or OpenFGA still leaves enforcement in the code. Classic edge authorization moves enforcement into the gateway, but business rules live in the APIs, so every API change means asking the gateway's owners: "it doesn't work, it doesn't scale". His modern edge pattern lets each team set the gateway's rules declaratively, and the gateway issues a signed internal token, which he calls an authorization contract, that the receiving service checks. He says a sidecar decision point can do the same, and that few tools support the modern edge approach yet. Which pattern fits depends, he says, on where the policy and its data live, and in practice the patterns are combined.
Five places to put authorization in microservices, and why Drus says the classic gateway one does not scaleBI-FIND-2026-1487
Asked in November 2025 how an assistant could hand its access to downstream agents, South said on-behalf-of flows and delegated authority existed as proposals "but they're not widely supported". Separately, he told of a developer whose MCP-connected agent, running with his super-admin privileges, took down Netlify's core website, and warned that "just because you as a person are extremely trustworthy, that does not necessarily mean Claude code using your permissions is extremely trustworthy". He also expected dynamic client registration in MCP to persist by path dependency even after client ID metadata arrived.
WorkOS's Tobin South, in November 2025: agents will get a new kind of account, not a new identity standardBI-FIND-2025-0085
Deriving an agent's permissions from the user's request, and enforcing them outside the model, stopped every injected instruction in this paper's test and stopped most of the legitimate work with it. In strict mode, across 240 runs on the AgentDojo benchmark, no attack succeeded, 92.2 percent of tool calls were denied and the agent completed 33.3 percent of its tasks. In permissive mode it completed 65.8 percent and three attacks got through, each because the parsed permission was a wildcard, write to any calendar or send to any address. The failure the method has is the one it was built to remove: authority granted more broadly than the task needed.
The IBAC paper's strict mode blocked every prompt injection, and its agent finished only a third of its tasksBI-FIND-2026-0310
Cited in 5 briefings
Every briefing that discusses this term, newest first.
- Five places to put authorization in microservices, and why Drus says the classic gateway one does not scale
Video · · OWASP Foundation · Uploaded October 21, 2026
- WorkOS's Tobin South, in November 2025: agents will get a new kind of account, not a new identity standard
Podcast · · CodingCat.dev Podcast · 5.10 - How to Secure AI - Identity Access Management for Agents · Nov 21, 2025
- The IBAC paper's strict mode blocked every prompt injection, and its agent finished only a third of its tasks
Article · · ibac.dev · Jordan Potti · Mar 3, 2026
- Huawei researchers add agent delegation to an existing permission model and prove each action traces to a human
Paper · · arXiv · Amjad Ibrahim et al. · Jun 2, 2026
- Auth0's AgentCore guide says prompts alone are not enough, then lets the prompt decide when to ask a person
Article · · Auth0 Blog · Kapil Patil · Feb 10, 2026
Related terms
What appears alongside OpenFGA in the same briefings.