megachangelog
Improvement0.0.91

SDK v0.0.91 – Enhanced error handling, provider compatibility, and model catalog updates

This release improves finish reason handling to prevent false completions, increases MCP server initialization timeout from 3s to 10s for better reliability on Windows, fixes custom Anthropic endpoint compatibility by gating server-side options, and enhances reasoning model support across adapters. It also adds optional API key support for various providers, opt-in Langfuse tracing for BYOK providers, and refreshes the model catalog with new providers and default model updates.

  • Responses that end with a missing or unrecognized finish reason are no longer treated as successful completions. AgentModelFinishReason gains unknown (the AI SDK adapter maps unified other and missing reasons to it instead of stop). Without tool activity, the agent keeps the partial response and continues once with a hidden user message; a second unknown finish fails the run. Queued user messages are now consumed at every request boundary, including the first iteration
  • Stdio MCP servers get a 10s default initialize budget (was 3s). npx/uvx launchers through cmd.exe on Windows routinely took 3-6s and were silently dropped from the session. An explicit timeout still overrides it
  • The Anthropic provider sends the server-side refusal fallbacks option only to api.anthropic.com. Custom endpoints such as Azure AI Foundry rejected it with a 400. New isOfficialAnthropicEndpoint helper
  • The portable reasoning level for the cline and openai-compatible adapters snaps to the model's advertised effort levels. Kimi K3 (low/high/max) was sent medium and minimal, which it rejects
  • OpenAI-compatible providers (generic, vercel-ai-gateway, user-defined ids) write providerOptions under the camelCase alias only, which stops the per-chunk providerOptions key 'openai-compatible' deprecation warning from @ai-sdk/openai-compatible
  • New apiKeyOptional provider fact (GatewayProviderMetadata.apiKeyOptional, resolveProviderApiKeyOptional, ProviderListItem.auth.apiKeyOptional), declared on openai-compatible, ollama, lmstudio, litellm, vertex, bedrock, and user-added endpoints from providers.json/models.json
  • Rules: new listEnabledRulesFromRecords / loadRulesForSystemPromptFromRecords exports. The combined user-instruction service now orders rules by name, matching the single-workspace path
  • Opt-in Langfuse tracing for BYOK providers (CLINE_LANGFUSE_ALL_PROVIDERS), plus operator tags, metadata and environment via CLINE_LANGFUSE_TAGS, CLINE_LANGFUSE_METADATA, LANGFUSE_TRACING_ENVIRONMENT
  • Hook discovery telemetry handles Windows path separators
  • Refreshed the model catalog. The Cline free list adds Solar Mini 4 and drops DeepSeek V4.1 Flash and space-bunny-alpha. Adds the Engy provider. Default models change for AKI.IO, Blue Claw, CoralBricks, CrossModel (GPT-6.1 Sol), DevPass, LLM Gateway, Mistral, NanoGPT, Requesty (Mistral Large 4), Neon (Claude Opus 5.5), Subconscious, and The Grid AI

Full Changelog: sdk/sdk/v0.0.90...sdk/sdk/v0.0.91

sdkprovidersmodel-catalogmcp-serversapi-compatibilitytracing

Source: original entry ↗

More from Cline

Follow Cline to get its new changes in your feed and email digest.

Fix0.0.45

Desktop v0.0.45

SSH remotes and environment connections now work correctly, with improved branch checking performance. Long model names display fully when space allows, reasoning mode works on additional models, and the model catalog has been refreshed with new defaults including Claude Haiku 5.5.

sshdesktopmodelsstabilityperformance
Improvement3.0.70

CLI v3.0.70

Multiple bug fixes and improvements including fixes for GPT-6 Astra and Claude Fable 5 error handling, MCP server input reading, settings migration, and CLI command validation. Enhanced header preservation for MCP add command, improved error messaging for invalid options, and refreshed model catalog with Claude Haiku 5.5 support.

clibug-fixesmcpmodelsvalidation
Improvement0.0.92

SDK v0.0.92

Unified AI SDK to single release line (ai 7.0.120 with @ai-sdk/* packages on provider-utils 5.0.50), improved reasoning/thinking handling for GPT and Claude models that don't support these features, fixed EPIPE handling in stdio MCP clients and removed Winston's uncaught exception handler from SAP AI Core provider, added parseMcpHeaders export to @cline/core, and refreshed model catalog including new Claude Haiku 5.5.

sdkai-modelsmcpstabilitycatalog
See all Cline changes →