Commit graph

4 commits

Author SHA1 Message Date
grs
7c1998db25
feat: fine grained access control policy (#2264)
This allows a set of rules to be defined for determining access to
resources. The rules are (loosely) based on the cedar policy format.

A rule defines a list of action either to permit or to forbid. It may
specify a principal or a resource that must match for the rule to take
effect. It may also specify a condition, either a 'when' or an 'unless',
with additional constraints as to where the rule applies.

A list of rules is held for each type to be protected and tried in order
to find a match. If a match is found, the request is permitted or
forbidden depening on the type of rule. If no match is found, the
request is denied. If no rules are specified for a given type, a rule
that allows any action as long as the resource attributes match the user
attributes is added (i.e. the previous behaviour is the default.

Some examples in yaml:

```
    model:
    - permit:
      principal: user-1
      actions: [create, read, delete]
      comment: user-1 has full access to all models
    - permit:
      principal: user-2
      actions: [read]
      resource: model-1
      comment: user-2 has read access to model-1 only
    - permit:
      actions: [read]
      when:
        user_in: resource.namespaces
      comment: any user has read access to models with matching attributes
    vector_db:
    - forbid:
      actions: [create, read, delete]
      unless:
        user_in: role::admin
      comment: only user with admin role can use vector_db resources
```

---------

Signed-off-by: Gordon Sim <gsim@redhat.com>
2025-06-03 14:51:12 -07:00
Derek Higgins
2e807b38cc
chore: Add fixtures to conftest.py (#2067)
Add fixtures for SqliteKVStore, DiskDistributionRegistry and
CachedDiskDistributionRegistry. And use them in tests that had all been
duplicating similar setups.

## Test Plan
unit tests continue to run

Signed-off-by: Derek Higgins <derekh@redhat.com>
2025-05-06 13:57:48 +02:00
Sébastien Han
79851d93aa
feat: Add Kubernetes authentication (#1778)
# What does this PR do?

This commit adds a new authentication system to the Llama Stack server
with support for Kubernetes and custom authentication providers. Key
changes include:

- Implemented KubernetesAuthProvider for validating Kubernetes service
account tokens
- Implemented CustomAuthProvider for validating tokens against external
endpoints - this is the same code that was already present.
- Added test for Kubernetes
- Updated server configuration to support authentication settings
- Added documentation for authentication configuration and usage

The authentication system supports:
- Bearer token validation
- Kubernetes service account token validation
- Custom authentication endpoints

## Test Plan

Setup a Kube cluster using Kind or Minikube.

Run a server with:

```
server:
  port: 8321
  auth:
    provider_type: kubernetes
    config:
      api_server_url: http://url
      ca_cert_path: path/to/cert (optional)
```

Run:

```
curl -s -L -H "Authorization: Bearer $(kubectl create token my-user)" http://127.0.0.1:8321/v1/providers
```

Or replace "my-user" with your service account.

Signed-off-by: Sébastien Han <seb@redhat.com>
2025-04-28 22:24:58 +02:00
Ashwin Bharambe
01a25d9744
feat(server): add attribute based access control for resources (#1703)
This PR introduces a way to implement Attribute Based Access Control
(ABAC) for the Llama Stack server.

The rough design is:
- https://github.com/meta-llama/llama-stack/pull/1626 added a way for
the Llama Stack server to query an authenticator
- We build upon that and expect "access attributes" as part of the
response. These attributes indicate the scopes available for the
request.
- We use these attributes to perform access control for registered
resources as well as for constructing the default access control
policies for newly created resources.
- By default, if you support authentication but don't return access
attributes, we will add a unique namespace pointing to the API_KEY. That
way, all resources by default will be scoped to API_KEYs.

An important aspect of this design is that Llama Stack stays out of the
business of credential management or the CRUD for attributes. How you
manage your namespaces or projects is entirely up to you. The design
only implements access control checks for the metadata / book-keeping
information that the Stack tracks.

### Limitations

- Currently, read vs. write vs. admin permissions aren't made explicit,
but this can be easily extended by adding appropriate attributes to the
`AccessAttributes` data structure.
- This design does not apply to agent instances since they are not
considered resources the Stack knows about. Agent instances are
completely within the scope of the Agents API provider.

### Test Plan

Added unit tests, existing integration tests
2025-03-19 21:28:52 -07:00