- Replace tidwall/gjson and tidwall/sjson with github.com/MoeForge/moego/ojson
and encoding/base64 with github.com/MoeForge/moego/base64 across the main
module and examples
- Introduce cross-method *ojson.Doc request rewriting pipelines across
thinking, signature, client/codex, runtime/executor/helps, pluginhost, and
all provider executors with FinalizePayloadDoc as the final semantic barrier
- Convert all consecutive JSON mutations (>= 2 Set/SetRaw/Delete operations)
to batch edits on ojson.Doc
- Upgrade read-only byte paths to ojson.GetBytesNoCopy,
ojson.GetManyBytesNoCopy, and util.ParseGJSONBytesNoCopy with explicit
strings.Clone ownership protection across streaming and persistent state
- Use base64.StdEncoding.ValidateString for zero-allocation signature checks
and base64.EncodeDataURL for single-allocation data URL construction
- Add AuthProviderEntry and AuthProviderEntries in sdk/pluginapi, forward entries over auth.identifier RPC, and send multi_auth_providers in lifecycle requests.
- Normalize per-entry OAuth display metadata with plugin metadata fallbacks in RegisteredPluginInfo while preserving legacy fields when entries are omitted.
- Expose HTML-sanitized oauth_provider_entries in management plugin list responses alongside oauth_provider and oauth_providers.
Closes: #7170
Bind modelStreamBridge entries to the caller's plugin ID and callback instance
so host.model.stream_read and host.model.stream_close reject cross-plugin or
cross-instance access. Also close model streams when callback instances or
plugins are torn down.
Fixes#7159
Plugin executor streams whose output format equaled the requested format bypassed stream translation, so SSE-framed chunks ("data: {...}" plus "data: [DONE]") reached the OpenAI/Gemini handlers unchanged. Those handlers add their own "data:" prefix and terminal marker, producing a doubled "data: data:" prefix and a duplicated [DONE] on the wire.
Built-in executors always run TranslateStream, and the registered same-format translators strip that framing. Route same-format plugin executor streams through the same translation instead of returning the channel or payload unchanged. Formats without a same-format translator keep the existing behavior via the TranslateStream fallback, which returns the payload unchanged, and the OpenAI Responses usage-details fixup still runs after translation.
Co-Authored-By: Claude Code <noreply@anthropic.com>
- Add Capabilities.AuthProviders and MultiAuthProvider (Identifiers() []string) in sdk/pluginapi, bump sdk/pluginabi.SchemaVersion to 10, and forward AuthProviders in sdk/pluginhost
- Extend internal/pluginhost to register, expose, and route multiple OAuth login identifiers per plugin across StartLogin, PollLogin, RefreshAuth, ParseAuths, and ExchangeCallback, while preserving per-credential oauth_provider/portal attribution for shared credential providers
- Expose oauth_providers alongside primary oauth_provider in RegisteredPluginInfo and GET /v0/management/plugins, and support target-credential re-login (StartOAuthV8) across sibling portals
Closes: #7126
- Add ProviderModels, Registrations, MultiProviderModels, and SchemaVersionMultiProviderModels so a single plugin can register model catalogs across multiple providers while preserving single-provider compatibility.
- Track per-plugin provider sets and multi-provider model registrations in the plugin host for catalog aggregation, per-auth model discovery matching, and targeted unload cleanup.
Closes: #7115
Existing request interceptors wrap the full request body in a JSON RPC envelope with Base64 encoding and skip subsequent duplex WebSocket business frames, causing ~3.33x memory and CPU amplification when rewriting small fields such as Codex `<timezone>` and `<current_date>` in large contexts.
- Add schema v8 raw request body patch methods (`request.patch_body_before` and `request.patch_body_after`) and capabilities (`request_interceptor_patch_body_before/after`, `request_body_patcher_before/after`) that pass raw `Body []byte` directly across the C ABI without JSON/Base64 request wrapping
- Add `RequestBodyEdit`, `RequestBodyPatchResponse`, and `RequestBodyPatcher` in `sdk/pluginapi` with single-pass host splicing, bounds/overlap validation, `json.Valid` verification, no-op original body reuse, and sequential multi-plugin chaining
- Eliminate redundant full-body slice clones in handler/conductor request interceptor application paths
- Run before-auth and after-auth request interceptors on subsequent duplex WebSocket `response.create`, `response.append`, and `response.steer` messages
Closes: #7072
- Let schema v7+ frontend auth providers opt out of receiving request bodies via frontend_auth_provider_omit_body / frontend_auth_omit_body so metadata-only auth plugins skip reading, decoding, cloning, and marshaling the request body during authentication.
- Avoid cloning the buffered request body slice before JSON serialization when invoking RPC-backed frontend auth providers that still inspect the body.
Closes: #7071
- Bump plugin SchemaVersion to 7 and let request interceptors declare per-phase body opt-out via request_interceptor_omit_body_before/after capabilities, so metadata-only plugins no longer pay for body cloning, Base64 encoding, and cross-ABI copies.
- Skip the body clone in interceptRequest for schema v7+ plugins that opt out of the current phase while preserving chain state, blind body replacement, and termination semantics.
Closes: #6998
- generate a unique selection_id for each handled Scheduler.Pick decision and freeze it into the execution context and UsageReporter
- propagate SelectionID through core usage records and pluginapi.UsageRecord while clearing stale IDs on unhandled, rejected, Home, and credits fallback paths
Closes: #6988
- Expose the opaque caller scope handle on request interception structs and accept it on HostModelExecutionRequest and internal model execution requests so replayed executions retain caller-scoped metadata without exposing client API keys.
- Resolve Claude MCP tool alias secrets from caller_scope metadata before falling back to hashed context API keys so plugin replays reproduce the original upstream tool names and derived session identities.
Closes: #6961
- Defer context cancellation in TestCancelableConnClosesUnderlyingTransportWhenCloseRacesCancel to satisfy go vet lostcancel across all control flow paths.
- Remove the unused streamBridgeStream.close wrapper left behind after the closeWithError refactor in internal/pluginhost.
- Annotate shared "gin" string context keys in TestIssue6825SDKCallerReplayIsolation with nolint:staticcheck directives.
Closes: #6953
- Apply canonical thinking normalization, suffix stripping, and plugin model metadata fallback in prepareExecutorCall across Execute, ExecuteStream, and CountTokens.
- Resolve plugin-registered ThinkingApplier by provider key before format fallback and validate unknown thinking suffixes on non-native-Responses models.
- Run thinking normalization before NewPayloadFinalizer so user payload rules act as the final host-side mutation.
Closes: #6940
- Apply helps.NewPayloadFinalizer once in prepareExecutorCall after translating requests to the plugin input format so Execute, ExecuteStream, and CountTokens honor configured payload rules.
- Match rules on the suffix-free model name while preserving the plugin request model, canonicalize source format aliases, and use the request root for Antigravity envelopes.
Closes: #6939
- Add optional retry_after_seconds to pluginabi.Error and preserve it in rpcError and structured stream error callbacks.
- Expose RetryAfter and SafeResponseHeaders on rpcError with duration overflow clamping so auth cooldowns and HTTP responses honor upstream retry hints.
- Allow auth.SafeResponseHeaders to extract Retry-After headers from errors implementing SafeResponseHeaders.
Closes: #6937
- Strip standard and credential-like headers from plugin HTTP callback requests whenever a redirect chain changes scheme, host, or port across both default and wire-profile client paths.
- Normalize default HTTP/HTTPS ports when comparing redirect origins and delete matching header map keys directly to handle non-canonical header casing across multi-hop redirects.
Closes: #6936
- Add DecodedBody to FrontendAuthRequest and populate it on routes decoded by ReadRequestBody while preserving raw Body and Headers.
- Enforce pre-auth zstd window, decoded size, coding layer, and concurrency bounds, keeping decode slots held until guarded RPC calls complete.
- Add regression tests covering zstd decoding, route filtering, size limits, RPC marshaling, and canceled RPC slot lifecycle.
Closes: #6886
- Add a loadCleanupDoneHook callback in pluginhost.Host so canceled plugin load tests wait until the in-flight load token is cleared after client shutdown.
- Drain dataChan after canceling the context in TestHandlerPluginExecutorPublishesStreamCancellation before closing the chunk channel to avoid racing ctx.Done() against normal stream EOF.
Closes: #6885
- Capture the post-translation, post-payload-rule service tier on UsageReporter and propagate EffectiveServiceTier through SDK usage records, plugin usage adapters, and Redis usage queue payloads.
- Ensure executors record translated request metadata after final payload configuration rules so server-side service_tier overrides are accurately reported alongside client and response tiers.
Closes: #6873
- Re-run AccessManager authentication before each responses websocket turn and duplex continuation using a cloned POST request with handshake headers stripped.
- Add AuthErrorCodeDenied and NewDeniedError in sdk/access so terminal provider refusals preserve their HTTP status and stop the provider chain.
- Add DenyStatusCode and DenyMessage to pluginapi.FrontendAuthResponse and map explicit plugin denials in accessAdapter.Authenticate.
Closes: #6843
- Define MethodHostAuthModels and HostAuthModels request/response types in pluginabi and pluginapi so plugins can query registered models by credential.
- Implement callHostAuthModels in pluginhost to return registry models for all or filtered credentials in the auth manager, including config-backed API keys.
Closes: #6815
- Snapshot plugin scheduler and preferences atomically to prevent deadlocks and data races during auth picking.
- Release host lock before notifying executor manager during adapter registration to prevent lock contention.
Closes: #6744
- Update `authIDForPath` and path resolution logic to consistently handle relative paths within the auth directory.
- Add robust relative path checking to prevent path mismatch issues during credential saves and model lookups.
Closes: #6722
Stop RegisterModels from publishing a synthetic plugin client for
plugins without an executor, which advertised their models without
any routable credential and bypassed per-auth model exclusions.
Retain the static catalog only for executor arbitration and
per-auth registration, so public availability now follows active
credentials and their exclusions. Reserve native provider models
before plugin arbitration to preserve native priority, and remove
the now unused model client commit path.
- Add `DisableType` support to plugin host adapters and registry mappings.
- Route disabled thinking to `between_tools` for models like Claude Sonnet 5.5.
- Fall back to low-effort adaptive thinking for models that reject disabled states.
Closes: #6698
- Add support for handling multiple auth records from a single file.
- Preserve plugin-disabled states and prioritize existing auth states when updating.
- Unregister stale auths and associated models to maintain consistency.
- Introduce dynamic auth ID assignment and file-based deduplication for better reliability.
- Improve metadata normalization, priority handling, and file path matching.
Closes: #6551
Expose quota_plugin_id on credential entries using the host's two-pass selection precedence, excluding fused plugins and selecting healthy alternatives.
Stop advertising declarative quota_probe support in v8 while preserving v0 behavior. Add regression tests for selection precedence, plugin fusion, and version-specific quota metadata.
Fixes#6519
- Inject API key, access provider, and native key flags into the request context during authentication
- Add `IsNativeKey` and `AccessProvider` fields to usage records, plugin API types, and redis queue payloads
- Provide context helper functions and automatic fallback resolution in the usage manager and reporter
Closes: #6509
- Automatically detach and shut down plugins that are disabled or unselected during configuration reload
- Unload all active plugins when the plugin system is globally disabled
Closes: #6471
An attachment a translator could not represent was dropped without a word, and
a turn whose only content was that attachment went upstream as an empty
messages, parts, or input array. The model then answered as if nothing was
attached, or the upstream rejected a request that never named the attachment.
A base64 document now reaches OpenAI as a file part. A file the target cannot
receive, such as container_upload, an OpenAI file_id, or input_audio sent to
Claude, is reported by name so the route answers 400 instead of forwarding an
empty turn. Text beside it still travels.
One rule, ErrIfNothingLeft, decides when to refuse, and RegisterCheckedRequest
carries the refusal on the request envelope. Every executor that translates
checks it before calling upstream. The exported converters keep their []byte
signature, so no caller or sibling test had to change.
- Add `FlushResponse` to prefer error-aware flushing and unwrap wrapped response writers
- Bypass Gin's void `Flush` while maintaining header accounting via `WriteHeaderNow` to keep transport errors observable
Closes: #6332
Check plugin model declarations without executing discovery. Share one per-account refresh worker across registration and periodic triggers, retaining one latest queued snapshot with its publication epoch. Add ownership, coalescing, and stale-publication regression tests.
- Add `MethodHostRoutingResetCooldown` RPC callback to plugin ABI and host dispatcher
- Introduce `HostRoutingResetCooldownRequest` and `HostRoutingResetCooldownResponse` in plugin API
- Reset credential quota and cooldown routing state in memory without persisting to token files
Closes: #6274
- Fall back to top-level token usage when Codex response usage contains only service tier metadata
- Propagate parsed service tier while retaining top-level token counts
- Avoid early return during stream usage observation for chunks with zero token counts
Closes: #6261
- Handle apply_patch custom tool call streaming deltas and snapshots for Responses API translation
- Track and resolve step indexes, function call identities, and error states across stream events
- Propagate apply_patch tool input errors and upstream failure states in stream and non-stream responses
- Add `Path` field to model execution requests and plugin interceptor responses
- Propagate custom request paths through execution metadata
- Allow image models during execution when entry or exit protocol is `openai-image`
Closes: #6196
- Parse `disabled` field from auth file metadata when listing and registering auth files
- Update auth entry status to disabled when `disabled` flag is set to true
- Generalize boolean parsing helper to support common boolean metadata fields
Closes: #6153