Playbooks · Coding tasks
Zalando's broker hands agents OAuth safely
Written November 8, 2026
Zalando open-sourced an inspectable OAuth delegation and consent broker for AI agents. What it brokers, and what it doesn't solve.
The problem it answers
Handing an agent an OAuth token means handing it the keys to a third-party service, with no inspectable record of what the agent was allowed to do with them. The token travels with the agent, the agent calls whatever it calls, and the only record is whatever the agent chose to log. Zalando's incubator open-sourced the Agentic Identity Broker to move that arrangement somewhere auditable: the agent never holds the user's provider token at all.
The broker is MIT-licensed and written in Go. Its repository describes it as managing delegation chains for on-behalf-of flows, bridging distinct OAuth 2.0 infrastructures, and providing a token vault. Zalando says it runs the broker in production internally for several MCP servers.
What it brokers
The delegation and consent model centres on three objects:
- Permission sets. Named, administrator-defined, human-readable groups of OAuth 2.0 provider scopes. A permission set names what an agent may do, in terms an administrator chose, rather than passing raw provider scopes through.
- Grants. Revocable, optionally time-limited delegations from a user to an agent. A grant is the record of who delegated what to whom, and it can be taken back.
- User sessions. Encrypted provider tokens held for one user and one third-party service, shared by all agents that user has delegated to. The tokens live in the broker's vault, not in any agent.
The call path runs through a gateway. The agent sends its own user-agent token only to the trusted gateway — Zalando's post names agentgateway — and the gateway validates that token and asks the broker to exchange it for an access token issued by the target's OAuth 2.0 infrastructure, refreshing it if needed. The exchange follows RFC 8693, OAuth 2.0 Token Exchange. The agent calls the tool with the exchanged token and never sees the provider token.
Consent is a separate surface from scopes. The broker's consent screen names a business capability, such as reading repositories, instead of presenting a raw scope string — Zalando's reasoning is that provider scope vocabularies are inconsistent and often too coarse for a person to evaluate. A user can reduce an agent's effective permissions on that screen when they are not comfortable granting everything it asks for. The broker also supports central tool approvals, so a tool call can be checked before an email is sent or a document shared, with a central audit trail.
What it does not solve
- A broker records consent; it does not make a compromised agent safe. Delegated access is the intersection of the user's permissions and the agent's permissions. An agent that is tricked into misusing a permission it legitimately holds still misuses it — the broker's record shows what was allowed, not whether the use was wise.
- It does not replace an identity provider. It reuses existing identity investment and can either mint tokens itself or proxy an existing upstream OAuth 2.0 authorization server. A team with no identity provider still needs one.
- It assumes a person delegates to an agent. The current features cover on-behalf-of flows, where a human actor delegates. Headless agents acting without a user present are on the roadmap, not in the broker today.
- It is early software. The project describes itself as v0.1 and in active development, coding in the open to gauge its approach. The production claim is Zalando's own statement about its internal use, not an independent review.
What to do with it
- Read the consent model before the code. The decision that matters is what permission sets your administrators define and what the consent screen shows users. A broker with permission sets as broad as the raw scopes underneath adds a vault and a log, not control.
- Try it against one service first. The repository's quickstart runs the broker locally with Docker Compose. Connect one third-party service, delegate one agent, and read the audit log for one tool call before wiring it into a fleet of agents and servers.
- Check the audit trail is actually reviewed. The broker writes structured JSON logs for compliance and monitoring, and records grant lifecycle changes. A log nobody reads is the same as no log — decide who reads it and when before the first delegation.
- Do not treat it as a substitute for scoping the agent. Keep the agent's own permissions narrow. The broker limits what a token can do; it cannot limit what an agent asks for, and a grant that covers everything the user can do is a grant the broker will faithfully honour.
Limits
- This page reads the broker's repository README and Zalando's engineering post. It reports no install or run of the broker here, so it says nothing firsthand about setup effort, what breaks during configuration, or how the consent screen behaves with a real third-party service.
- The production-use and version statements are the project's own. No independent install or review exists in the sources read for this page.
- The mechanism described here is the documented design: permission sets, grants, user sessions, and gateway token exchange. How it behaves with a provider whose OAuth implementation differs from the standard is not covered by the sources read.
Related playbooks
- Open WebUI 0.13.3 puts agents in automations
- OpenCode Browse: agents use your real browser
- Recalc: agents inside spreadsheet cells
- Manage Git worktrees for parallel AI agents with Worktrunk
Sources
- Agentic Identity Broker — GitHub
- Agentic Platform: Open-Sourcing the Agentic Identity Broker — Zalando Engineering
Back to: Coding playbooks hub · Also: Audit apps that installed agent plugins for you · Check your Codex and Claude Code config for relay tampering · Lock down Claude: IP allowlist and sign-out