cline/sdk/CHANGELOG.md

Cline SDK Changelog

0.0.52

• Added checkpoints support to the agent runtime
• Added SAP AI Core provider support: stabilized provider setup, bundled provider auth, forwarded provider options to the gateway, aligned provider config, kept model filtering in clients, and added OCA legacy reasoning-effort handling
• Routed LiteLLM model fetches through the SDK and stopped unrelated models from being injected into the LiteLLM model list
• Preserved OpenRouter reasoning-disable semantics and included the session id for OpenRouter prompt caching
• Updated the ClinePass model list live, restored ClinePass models in onboarding, fixed ClinePass error mapping, and scoped the ClinePass URL to the CLI
• Threaded proxy/CA-aware fetch into the SDK inference path
• Persisted Bedrock settings to providers.json
• Repaired exposed provider auth routing and restored provider-request capture wiring lost in the SDK migration
• Added a connector configure path and moved the shared connector catalog into the shared package
• Normalized JSON-like tool inputs by schema and avoided a nullable editor old_text schema
• Batched outdated-read rewrites in MessageBuilder to preserve provider prefix caches
• Prevented an "ERROR: EMPTY CONTENT" message from appearing when an error occurs
• Added non-interactive command guidance to the agent
• Published SDK sourcemaps
• Refreshed the generated model catalog

0.0.51

• Fixed Z.ai model metadata not resolving correctly when using Z.ai models through the Cline provider; aliases now map to the right model metadata and user overrides are preserved

0.0.50

• Truncate every tool result by default (including MCP and custom tool output), with tightened MessageBuilder limits and tunable CLINE_MESSAGE_BUILDER_* env overrides, to keep provider requests within budget
• Cap assistant text in provider messages and count tool_use input toward the request budget; protect binary carrier blocks (not just images) from truncation
• Resolve tool names from tool_result when the paired tool_use is gone
• Add ClinePass provider support (built-in provider, error handling, format compatibility)
• Apply auto-approve toggles immediately in the agent runtime
• Harden parallel tool-call guidance in the system prompt and tool definitions
• Refresh the generated model catalog

0.0.49

• Reverted ClinePass recommended-models support, removing the clinePass field from the recommended models data

0.0.48

• Added ClinePass support and ClinePass models
• Added MCP server support to plugins
• Updated the recommended/fixed model list
• Encouraged parallel tool calls for faster task execution
• Capped tool output ingestion for bash commands and file reads to keep large output within context limits
• Added a bounded media budget for provider requests, plus generic provider-request capture
• Allowed ranged reads on large files
• Fixed apply_patch to fail when a hunk is skipped instead of silently dropping it
• Fixed run_commands to return captured stdout on failure and to coalesce split heredocs
• Fixed search tools to treat zero results as a successful result
• Fixed search output cap and bash executor follow-up issues
• Fixed disabled-reasoning handling for StepFun flash
• Fixed the Hugging Face URL
• Fixed Cline OAuth token formatting in provider config

0.0.47

• Added support for overriding the API base URL
• Enforced a production singleton Cline Hub so only one hub daemon runs, and a stale hub is respawned after an upgrade
• Allowed plugin chat commands to submit prompts to the agent
• Fixed truncation of structured tool operation result strings so oversized tool output stays within limits
• Stopped echoing the full command text in run_commands tool results

0.0.46

• Added support for configured agents as subagent tools
• Centralized OAuth management into the SDK
• Added Vertex GCP settings configuration
• Fixed the Azure Foundry API version for the CLI
• Fixed an error caused by disabled reasoning on Fable 5

0.0.45

• Added support for the Claude Fable 5 model
• Fixed MiniMax M3 thinking controls so they route correctly across gateways

0.0.44

• Added support for Vertex AI Application Default Credentials (ADC) with tool use
• Added a global auto-update setting for CLI startup updates
• Fixed empty message content replay for Bedrock
• Cleaned up the OpenAI Codex model list

0.0.43

• Added the Cline Hub web app for managing and monitoring agent sessions
• Added plugin uninstall support
• Added skills bundled with plugins, including grouping plugin skills in settings and rule contributions from sandboxed plugins
• Added support for global AGENTS rules
• Added Slack socket mode support and bound Discord sessions to individual message authors
• Synced the Fireworks AI model registry and updated the model catalog to current platform offerings
• Routed custom registered handlers through the agent runtime
• Added a CLINE_PLUGIN_IMPORT_TIMEOUT_MS environment override for plugin import timeouts
• Allowed a baseUrl field for Anthropic vendor-type providers
• Fixed SAP AI Core to use the AI SDK community provider
• Fixed the hub daemon to stay alive on runtime abort
• Fixed read-files tool input validation to use a union schema
• Fixed discovery of symlinked SDK skill directories
• Improved Cline provider migration
• Fixed OTEL variable bundling
• Added telemetry for run_commands timeouts

0.0.42

• Supports Bedrock bearer API keys, direct IAM credentials, AWS profiles, and the default AWS SDK credential chain
• Routes Z.AI GLM thinking through provider metadata while preserving generic thinking suppression for non-GLM Z.AI custom models