Goose sürüm notları
Goose aracının son 25 kararlı sürümü, en yenisi üstte. Bir şeyi kaldıran, adını değiştiren ya da kullanımdan kaldıran satırlar en başta toplanır. Her sabah GitHub sürümlerinden alınır; maddeler orijinal haliyle İngilizcedir.
Kırıcı değişiklikler
- 1.48.0
- Deprecated Gemini OAuth provider
- 1.47.0
- Deprecate Gemini OAuth provider
- 1.46.0
- Deprecate and remove ui/text TUI
Son sürümler
-
- Allowlist: require HTTPS and validate redirects, block on policy failure
- Allow callers to supply Goose session IDs
- Linux window icons restored in desktop app
- OpenRouter: resolve variant suffixes like :free and :nitro to the base model
- Use catalog_provider_id for vision lookup on custom providers
- Desktop: start the agent only once after a fork
- Scheduler: use recipe settings in scheduled runs
- Voice: use system trust for WebSockets
- Prompt: surface MiniJinja render errors in custom system prompts
- Providers: request summarized thinking from Claude models
-
- Automatic thinking level operation for improved model reasoning
- Resume failed or cancelled delegate subagents
- Subagent session link in UI with view-only subagent sessions
- Limit concurrent subagents with GOOSE_MAX_CONCURRENT_SUBAGENTS
- Report session status via OSC 7501 in CLI
- Windows NSIS installer for desktop
- IO Intelligence provider
- Provider-defined cache prices support
- Dangerous include history option
- Replace PCTX code mode with Boa JavaScript runtime
-
- AIgateway declarative provider
- Propagate W3C trace context from ACP prompts to MCP tool calls
- Surface provider x-request-id on responses and errors
- Honor supports_vision declared on custom provider models
- Use live model metadata for OpenAI reasoning and prefer the Responses API
- Cap auto-compaction at 225k tokens by default
- Improve askai bot relevance
- Adaptive thinking for Claude Sonnet and Haiku 5.5
- Reject malformed GOOSE_MODE across ACP providers
- Bedrock: don't retry non-context ValidationException on converse_stream
-
- Prompt for goose mode during first-time configure
- HTML session export with an interactive viewer
- Adaptive thinking for Claude Sonnet and Haiku 5.5
- Treat a client extension selection as the exact session set
- Support hyphenated GPT 6.1 Sol effort aliases
- Use a system Node.js in the Windows npx wrapper
- Show thinking when GOOSE_CLI_SHOW_THINKING is set to any value
- Price Databricks GLM aliases from the zhipuai catalog
- Read messages from config for tom
- Preserve the service total when folding cache input tokens for Bedrock
-
- Resizable Picture-in-Picture window for MCP Apps
- Fetch model metadata from models.dev live with bundled fallback
- Quoting action to append selected conversation text to composer
- Lean ACP-only Goose binary
- Model discovery API for GDK Provider
- Meta Muse Code provider
- Support for Opus 5.5, GPT-6-sol, GPT-6-luna
- Session rename command in CLI
- Show working directory in session remove command
- GDK agent loop runs on wasm32
-
- Live voice conversations in the desktop app
- Decisions provider crate with OpenRouter and Jev implementations
- Z.AI Coding Plan provider with streaming tool calls
- Recipe parameter limits enforcement (max 32 params, 200 select options, 128 KiB)
- OpenAI provider custom base URL support in SDK
- Support for Opus 5.5, GPT-6-sol, and GPT-6-luna models
- Cancel elicitation on incomplete CLI input
- Protect custom provider configuration files
- Prevent duplicate schedules from overwriting recipes
- Classify many-image dimension limit errors as context length-exceeded
-
- Route Desktop to state-machine loop via ACP prompt meta
- EUrouter as a declarative provider
- Connect external backend ACP socket to redirect-resolved backend URL
- GPT-live API support
- Operator allowlist for gateway pairing
- Opt-in terminal bell when a turn finishes or approval is needed
- Report real cause of external backend connection failures
- Scope Toolshim to custom providers
- Treat auto-compact 100% as disabled
- Validate recipes consistently in scheduler
-
- Revert model updates for patch release
- Roll back MCP default version selection update #
-
- Support GPT-6 Astra models
- Tool calling for goose-agent
- Kotlin callers can now configure Databricks AI Gateway path
- Require HTTPS for Snowflake connections
- Bind CLI session naming to trusted path
- Honor content capture opt-out in tracing
- Enforce subagent platform guards
- Show authoritative tool approval details
- Preserve permission revocations across managers
- Resolve UI freeze after renaming session
-
- Auto-updater for desktop application
- Linux ARM64 desktop release packages
- Load extensions in the background so the CLI prompt is immediately usable
- Title sessions by subject, not workflow step
- Interactive git branch indicator in chat bottom bar
- on_failure block for PreToolUse hooks
- Web-search and browser-use built-in skills
- Model-native audio transcription provider
- --with-extension can now name its extension
- Auto-focus chat input when user starts typing
-
- Kırıcı Deprecated Gemini OAuth provider
- TrustedRouter declarative provider
- OpenCode Zen gateway as a declarative provider
- Gondola as declarative OpenAI-compatible provider
- SayGM as declarative OpenAI-compatible provider
- Lynkr as a declarative OpenAI-compatible provider
- PleumRouter declarative provider
- Z.ai GLM-5.2 replaced with GLM-5.3
- Custom provider cost fields drive cost tracking
- Model-native audio transcription provider
- OpenAI GPT-5.6 (sol/terra/luna) routing via AWS Bedrock
-
- Kırıcı Deprecate Gemini OAuth provider
- Interactive git branch indicator in chat bottom bar
- Support for pre-registered OAuth clients for streamable_http extensions
- Show recently used models in chat footer model picker
- Isolate concurrent subagent notifications in summon
- Validate recipe parameter values
- Re-enable minijinja builtins so documented recipe filters work
- Send explicit thinking disabled to adaptive Claude models
- Handle error code for context length exceeded
- Include required annotations field on replayed output_text items
- Reject context commands for context-owning providers
-
- Kırıcı Deprecate and remove ui/text TUI
- Unrolled agent loop
- Cache-safe request assembly with append-only turn context and declared cache semantics
- Streaming shell output while commands run
- Per-message usage stats UI (tokens, cost, TTFT, tok/s)
- Per-message usage/cost tracking with derived session totals
- Track cache tokens for accurate cost reporting
- Markdown format option to session export
- Resizable sidebar with drag handle and persisted width
- Interactive menu for single select elicitations
- Session metadata on sidebar chat hover
-
- Stable agent event message identity
- Tab completion and switch provider for model command
- Support latest Gemini models
- Configurable GOOSE_DOCS_ROOT for air-gapped docs access
- Structured summary output with template rendering for compaction
- Allow disabling built-in skills
- Gate tool-call label enrichment on ACP client capability
- Restore dynamic model discovery and current Grok support
- Update DeepSeek model names to v4 API
- Forward Claude Code model picker selection to the adapter
-
- Arbitrary command execution in goose CLI via goose review (via git core.fsmonitor config) - GHSA-r5pp-p5r8-466r (credit to @0xmagic0)
- Support for latest Gemini models
- Pass working_dir to the Stop hook context
- Declarative Sakana AI provider for OpenAI-compatible Fugu API
- Model interactions viewer restored
- Chunk command-classifier input with overlapping windows for security
- Search filter for provider grid
- Muse Spark 1.1 support via Meta Models API
- GPT-5.6 support
- Reconnect desktop ACP sessions after sleep and connection loss
-
- Reconnect desktop ACP sessions after sleep and connection loss
- Per-message usage stats UI (tokens, cost, TTFT, tok/s)
- Per-message usage/cost tracking with derived session totals
- Group chat sessions by project in nav panel
- Show project names in directory switcher dropdown
- OllamaCloudProvider with dynamic model discovery and context limits
- Delete support for custom apps from Apps UI
- --edit session flag to edit conversation before forking
- iFlytek Spark and Astron MaaS providers
- Fireworks AI declarative provider
-
- OllamaCloudProvider with dynamic model discovery and context limits
- Per-message usage stats UI (tokens, cost, TTFT, tok/s)
- Per-message usage/cost tracking with derived session totals
- Group chat sessions by project in nav panel
- Show project names in directory switcher dropdown
- Provider bindings MVP to goose-sdk with Python wheel publishing
- AskAI bot hints for followups
- Resolve provider default model instead of leaking sentinel "current"
- Return effective context limit from /model-info
- Recover malformed tool calls from GLM/Minimax models instead of rejecting them
-
- iFlytek Spark and Astron MaaS providers
- --edit session flag to edit conversation before forking
- Fireworks AI declarative provider
- OpenRouter request parameters support
- French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, and Traditional Chinese desktop locales
- Spanish (es) locale support
- Korean localization for desktop app
- Japanese locale support
- Hindi desktop locale
- Language selection feature in desktop
-
- Language selection in desktop app
- French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, Traditional Chinese desktop locales
- Korean localization for desktop app
- Japanese locale support
- Spanish (es) locale support
- Hindi desktop locale
- /status slash command in CLI
- /model slash command for session model switching
- /goal command for agent self-evaluation before finishing
- goose review local code review command
-
- ACP method for managing recipes and UI changes
- Use ACP to manage global config and session extensions
- Recipe handling in new_session
- /status slash command in CLI
- ACP new session on desktop (non-recipe sessions)
- Typed request/response for add, get session extensions in ACP
- Option to disable automatic update downloads
- Change cwd using ACP and refactor extension in bottom menubar
- Track cache tokens for accurate cost reporting
- Load session using ACP
-
- Opt-in ACP last message snippets for sessions
- Unified OTLP logging schema for cross-tool detection
- Canonical thinking modes across providers
- Custom ACP method to get session info
- ACP secret key support at goose serve endpoint
- Surface Anthropic stream refusals as visible errors
- ACP search session support in Desktop
- Steering messages with ACP
- ACP list sessions with keyword and type filter
- ACP thinking effort config option
-
- xAI SuperGrok OAuth subscription provider
- Replay ACP images on session load
- Expose raw provider supported models over ACP
- ACP session system prompt setter
- Scholar Sidekick MCP extension
- tui feature flag to gate the TUI command
- CLI to list skills with token counts
- Perplexity as a declarative OpenAI-compatible provider
- Alibaba (Qwen via DashScope) declarative provider
- Configurable tool output size limit via GOOSE_MAX_TOOL_RESPONSE_SIZE
-
- TUI command on goose-cli
- TUI diff viewer
- goose review local code review command
- /goal command for agent self-evaluation before finishing
- Hooks system for PreToolUse denial and extensibility
- Slash commands (built-in, skill, recipe) in ACP server
- Open-plugins generalization + skills
- Summon subagent instructions
- Unified thinking effort control across all providers
- NEAR AI Cloud provider
-
- Hooks system for extensible pre/post tool execution
- PreToolUse denial hook support
- /goal command for agent self-evaluation before finishing
- goose review local code review command
- Diff viewer in TUI
- Slash commands (built-in, skill, recipe) in ACP server
- Unified thinking effort control across all providers
- Open-plugins generalization + skills
- Summon subagent instructions
- Proactive OAuth token refresh to avoid re-auth on every session
-
- Build non-vulkan linux variants using ubuntu 22.04
Diğer ajanlar
Claude CodeCodex CLIGemini CLICopilot CLIopencodeHermes AgentQwen CodeKimi CLIMistral VibeCrushAiderCline Bütün ajanlar →
Goose aracını diğer ajanların yanında çalıştırın, hangisinin sizi beklediğini görün: AgentVera hepsini Mac, Windows ve Linux’ta tek pencerede, telefonunuzda da tutar.
Ajanlarını bir masaya topla.
İndir, projeni seç, ilk ajanını aç. Kurulu CLI’ların hazır.