Commit Graph
8 Commits
Author SHA1 Message Date
VarshaandJohn Myers 0803c4aa4c refactor(policy)!: remove NetworkBinary harness field (#3222)
* refactor(policy)!: remove NetworkBinary harness field

Closes #3054

Signed-off-by: Varsha Prasad Narsing <varshaprasad96@gmail.com>

* test(policy): preserve unknown fields during migration

Signed-off-by: Varsha Prasad Narsing <varshaprasad96@gmail.com>

* fix(policy): preserve advisor provenance during merge

Signed-off-by: Varsha Prasad Narsing <varshaprasad96@gmail.com>

* fix(policy): remove redundant network binary default

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(policy): record harness schema migration

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

---------

Signed-off-by: Varsha Prasad Narsing <varshaprasad96@gmail.com>
Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>
Co-authored-by: John Myers <9696606+johntmyers@users.noreply.github.com>
2026-09-11 16:49:14 +00:00
John T. Myers f4dc6be4b2 refactor(inference): remove managed inference routes (#3195)
* refactor(inference): remove managed inference routes

Closes #3172

Remove the inference route control plane, inference.local data path, built-in router crate, and SDK surface. Move inference workloads to explicitly imported provider profiles and native endpoints, with migration cleanup and updated tests and documentation.

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(policy): preserve alternate upstream isolation

Restore the provider policy activation guard so legacy OpenAI and Anthropic providers configured for alternate base URLs do not grant egress to the built-in public vendor endpoints.

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

---------

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>
2026-09-09 18:47:22 +00:00
John T. MyersandJohn Myers 07df822090 feat(providers): make profiles authoritative (#2962)
* feat(providers): make profiles authoritative

Closes #1988

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* docs(providers): move profiles into provider navigation

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* docs(providers): clarify provider attachment lifecycle

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(tui): scroll provider profile picker

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(providers): honor profile credential semantics

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(providers): prefer exact profile IDs

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(providers): harden authoritative profile adoption

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* test(oidc): align provider fixtures with profiles

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(providers): preserve authoritative profile lifecycle

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

---------

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>
Co-authored-by: John Myers <johntmyers@users.noreply.github.com>
2026-09-01 19:22:07 +00:00
John T. MyersandJohn Myers 0120535efc feat(proxy): bind static credentials to provider endpoints (#2510)
* feat(proxy): bind static credentials to provider endpoints

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* test(e2e): verify static credential endpoint isolation

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* docs(provider): explain static credential endpoint binding

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(e2e): use valid endpoint isolation fixtures

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* docs(provider): explain static credential endpoint binding

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>

* fix(credentials): preserve binding identity across rotations

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(proxy): enforce bindings across request lifecycle

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(proxy): close credential relay gaps

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(credentials): clarify binding failure behavior

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): hash selected provider profile scope

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(proxy): resolve credentials after request admission

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(credentials): clarify binding failure diagnostics

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(proxy): align single-route credential denials

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): harden endpoint-bound rotation

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): enforce identity and authority binding

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): snapshot provider environment atomically

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(e2e): include authority port in query proxy requests

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): close credential revocation gaps

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(proxy): explain authority mismatch diagnostics

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): enforce binding lifecycle invariants

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(provider): reject credential config collisions

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(network): capture credential scope atomically

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(network): distinguish origin and absolute targets

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(provider): isolate endpointless profile credentials

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(network): normalize IPv6 request authorities

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(credentials): clarify endpointless profile isolation

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* feat(policy): bind endpointless provider credentials

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): use current GCP placeholder revision

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(providers): explain policy credential bindings

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(credentials): cover endpointless fail-closed invariant

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(policy): expect ambiguity rejection at creation

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(server): authenticate rebased policy requests

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* refactor(proxy): share credential mismatch finding builder

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(credentials): cover malformed binding metadata

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(credentials): verify multi-key endpoint isolation

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(e2e): cover same-host credential path denial

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* docs(credentials): document serialized refresh contract

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* refactor(proxy): consolidate L7 log formatting

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* perf(credentials): precompile endpoint binding patterns

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* perf(credentials): share identity epoch revisions

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* test(proxy): require explicit request default ports

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(policy): validate SigV4 credential sources

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* fix(credentials): preserve endpoint bindings for credential handles

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

* feat(go-sdk): expose network credential bindings

Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>

---------

Signed-off-by: John Myers <johntmyers@users.noreply.github.com>
Signed-off-by: John Myers <9696606+johntmyers@users.noreply.github.com>
Co-authored-by: John Myers <johntmyers@users.noreply.github.com>
2026-08-10 19:19:45 +00:00
Russell Bryant aa483ecb9a feat(providers): AWS STS AssumeRole refresh strategy and aws-s3 profile (#1782)
Add gateway-managed AWS STS credential refresh (provider-v2, #1576). The
gateway calls sts:AssumeRole and writes three short-lived credentials
(AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_SESSION_TOKEN) to the
provider record; the proxy re-signs requests with SigV4. Adds the aws and
aws-s3 provider profiles and a declarative multi-output refresh model
(additional_outputs) so one AssumeRole co-mints all three credentials.

Signed-off-by: Russell Bryant <rbryant@redhat.com>
2026-07-16 13:28:01 -07:00
Jesse JaggarsandRussell Bryant f569a0ade6 feat(sandbox): proxy-side AWS SigV4 credential signing for CONNECT tunnels (#1638)
Signed-off-by: Jesse Jaggars <jjaggars@redhat.com>
Co-authored-by: Russell Bryant <russell.bryant@gmail.com>
2026-06-26 10:52:24 -07:00
Robert Sturla 48545cfb59 feat(sandbox): add GCE metadata emulator for Google Cloud (#1763)
* feat(core): add shared GCP constants module

Single source of truth for GCP naming: env var aliases, provider config
keys, token search order, and Vertex-specific env vars. Consumed by
openshell-server, openshell-providers, and openshell-sandbox.

- Add google_cloud.rs with metadata emulator host and loopback address
- Define PROJECT_ID, REGION, and SERVICE_ACCOUNT_EMAIL env var aliases
- Add provider config key constants for gcp provider implementations
- Define TOKEN_ENV_KEYS search order (SA token takes priority over ADC)
- Add Vertex-specific env vars for Goose and Claude Code SDK integration
- Add STATIC_CONFIG_KEYS as union of all alias arrays for env resolution
- Export module via openshell-core lib.rs

Signed-off-by: Robert Sturla <rsturla@redhat.com>

* feat(providers): add google-cloud and vertex provider plugins

Add GoogleCloudProvider and VertexProvider implementing inject_env to
project GCP config (project ID, region, SA email, metadata host) into
sandbox environment variables. Replace the inline Vertex AI env
injection in the server with the registry-based inject_env dispatch.

Also adds the google-cloud.yaml provider profile with SA JWT and ADC
OAuth2 credential refresh flows.

Signed-off-by: Robert Sturla <rsturla@redhat.com>

* feat(sandbox): add GCE metadata emulator for GCP

Add a loopback HTTP server on 127.0.0.1:8174 inside the sandbox
network namespace that emulates the GCE instance metadata API.
GCP client SDKs discover it via GCE_METADATA_HOST and obtain
credential placeholders that the proxy resolves to real tokens
at egress.

Add metadata_server module with MetadataHandler trait and
netns-aware TCP binding via std::thread (not spawn_blocking)
to avoid tokio pool namespace contamination
Add google_cloud_metadata module implementing the GCE metadata
API subset (token, project-id, email, scopes, service-accounts)
Add child_env_resolved() and gcp_token_response() to
ProviderCredentialState for GCP-aware credential projection
Wire metadata server into sandbox lifecycle before SSH handler
Collapse multi-line HTTP response format string into single line

Signed-off-by: Robert Sturla <rsturla@redhat.com>

* docs(sandbox): add GCP credentials documentation

Document the google-cloud provider setup for ADC and service account
flows, injected environment variables, metadata emulator behavior, and
network policy configuration for GCP APIs.

Signed-off-by: Robert Sturla <rsturla@redhat.com>

* feat(cli): support --from-gcloud-adc for google-cloud providers

Widen --from-gcloud-adc to accept google-cloud providers. The ADC
credential key is derived from the provider profile rather than
hardcoded per type, so future GCP provider types get ADC support by
declaring the right refresh metadata in their profile YAML.

Add ProviderTypeProfile::adc_credential() to find the ADC-compatible
credential from a profile's refresh metadata. Remove unused
VERTEX_AI_ADC_TOKEN_KEY and GCP_ADC_TOKEN_KEY constants.

Signed-off-by: Robert Sturla <rsturla@redhat.com>

---------

Signed-off-by: Robert Sturla <rsturla@redhat.com>
2026-06-23 15:05:07 -05:00
Adam Miller f061b1d923 feat(providers): add Google Vertex AI inference provider (#1568)
* feat(providers): add Google Vertex AI provider

Adds Vertex AI provider profiles, routing, credential refresh plumbing, CLI support, docs, and regression coverage. Keeps the related NETLINK_ROUTE seccomp allowance needed by Vertex client tooling that calls getifaddrs.

* docs: add Vertex AI sandbox usage for Claude Code and OpenCode

Cover the full end-to-end setup for running Claude Code and OpenCode
inside an OpenShell sandbox via inference.local with a Vertex AI backend:

- google-vertex-ai.mdx: add 'Use from a Sandbox' section with tabbed
  examples for Claude Code (--bare flag, no /v1 suffix) and OpenCode
  (/v1 suffix required). Add providers_v2_enabled prerequisite and
  --no-verify note for global region. Document policy proposals table
  covering metadata.google.internal (always blocked), downloads.claude.ai,
  and storage.googleapis.com.

- inference-routing.mdx: expand 'Use the Local Endpoint' section with
  tabbed examples for Claude Code, OpenCode, Python OpenAI SDK, and
  Python Anthropic SDK. Add notes explaining the /v1 path suffix
  difference between clients.

- supported-agents.mdx: update Claude Code and OpenCode rows to mention
  inference.local support and correct base URL requirements.

* fix: address vertex review findings

* test(sandbox): retry on spurious Ok in fork-exec ambiguity test

On arm64 under heavy CI load, the /proc fd scan in
find_socket_inode_owners can transiently miss the parent process's
socket fd entry, returning only the child as an owner. This causes
resolve_process_identity to return Ok (single owner, no ambiguity
check fires) instead of the expected ambiguous-ownership Err.

Extend the retry loop to also handle unexpected Ok results, mirroring
the existing retry for transient Err results. 10 retries at 50ms gives
a 500ms settling window, which is sufficient for procfs to stabilize
on loaded arm64 runners.

* fix: address vertex review regressions

* docs(router): clarify stream_response semantics for Vertex rawPredict routing

Document the three call sites of prepare_backend_request and their
stream_response values in a caller table:

- send_backend_request: false → :rawPredict (unary endpoint)
- send_backend_request_streaming: true → :streamRawPredict
- verify_backend_endpoint: explicitly false to probe the unary endpoint

Cross-reference the table from build_provider_url and
is_vertex_anthropic_rawpredict_route so the stream_response=true guard
in the suffix upgrade branch is understood in full context.

Also note that is_vertex_anthropic_rawpredict_route is a structural
predicate (model_in_path + anthropic_messages + :rawPredict suffix),
not a named-provider check, so any future provider with the same route
shape inherits the transforms automatically.
2026-06-02 10:45:50 -05:00