ArgoCD User Permissions
You set up argocd user permissions by mapping identity provider groups to role rules. These rules define which apps and clusters each person or team can touch. Many teams start by giving developers broad admin rights to speed up work. Broad roles let anyone delete production apps or change cluster settings by mistake. Tight permission rules keep clusters safe while letting teams ship code on time. How argocd user permissions work in practice The platform uses role-based rules to control what
Your coding agents run with your access.
Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.
You set up argocd user permissions by mapping identity provider groups to role rules. These rules define which apps and clusters each person or team can touch. Many teams start by giving developers broad admin rights to speed up work. Broad roles let anyone delete production apps or change cluster settings by mistake. Tight permission rules keep clusters safe while letting teams ship code on time.
How argocd user permissions work in practice
The platform uses role-based rules to control what users can do after login. Admins write policy lines inside a central config map on the host cluster. Each policy line names a subject, a resource type, an action, and an object. Subjects can be local accounts or group names from your identity provider.
Resource types include apps, clusters, repos, projects, and user accounts. Actions cover simple verbs like get, create, update, sync, override, and delete. Objects match specific project names and app names, often using wildcards for teams. You also set a default role for anyone who logs in without a group.
Most teams disable local users and link an external single sign-on system instead. When a developer logs in, the identity provider sends a token with group claims. The server reads those group claims and matches them to the policy rules. This flow lets you manage team membership inside your identity tool instead of cluster files.
Using projects to scope team boundaries
Global rules can grow messy when dozens of teams share one control plane. Projects give you a clean way to group related apps and set boundaries. A project restricts which source repos and target clusters its apps can use. It can also block certain cluster resource kinds from being deployed by a team.
You can bind roles directly inside a project instead of the global policy file. Project roles issue tokens or map to single sign-on groups for that specific scope. This design lets team leads manage their own app sync rules without admin help. Meanwhile, platform engineers keep full control over cluster Destinations and allowed source repos.
User Provisioning (SCIM) + ArgoCD Security: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
What native roles solve and where gaps remain
Native roles do a great job of splitting read and write duties across teams. You can give most engineers read-only views while letting a few leads trigger syncs. You can also lock production projects so only automated pipelines can push changes there. These rules stop casual mistakes inside the web interface and command line tool.
Still, static role mappings only check who logs in at the start of a session. Once a user has sync or update rights, those rights stay active all day. Static roles do not require peer review before a user runs a dangerous delete action. They also do not mask sensitive data like emails or phone numbers in live resource views.
During an incident, engineers often need temporary write rights to fix a broken app. With native config maps, an admin must edit policy files and revoke them later. Teams often forget to remove these temporary roles once the fire is out. Over time, permissions creep up and leave production clusters exposed to broad human access.
Check Kubernetes API requests and responses on the wire
Kubernetes RBAC decides who may call the API server. RBAC does not check response payloads or split listing Secret names from reading Secret values when a role grants get access. With hoop.dev, you run an open-source HTTP proxy, a sidecar for your Kubernetes API. The kubectl client changes only the server address it talks to while RBAC stays in place.
On each call, hoop.dev guardrails read the method, normalized path, and headers to refuse bad deletes or full Secret reads with a 403 error. On the way back, hoop.dev data masking walks JSON responses by key and masks matched values in Secrets, ConfigMaps, and pod logs. Every request and verdict is logged without storing bearer tokens. View the project on GitHub.
Frequently asked questions
How should we set the default role for new users?
Set the default role to read-only or an empty role with no rights at all. Then grant write or sync permissions only through specific identity provider group mappings.
Should we use local accounts or single sign-on groups for team access?
Use single sign-on groups for all human users so offboarding happens in one place. Keep local accounts disabled or limit them to automated pipeline jobs with tight scopes.
How can we prevent a team from deploying to the wrong cluster?
Put the team's apps inside a dedicated project with strict destination rules. Only list the exact cluster addresses and namespaces that belong to that specific team.
Related guides
In this post