spacehub

surshar / claude-code

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Built by Anthropic · github.com/anthropics

claude-code / CHANGELOG.md
GitHub Actionschore: Update CHANGELOG.md 5248fa0 · on Aug 8, 2025 History
532 lines · 18.4 KBPreviewCodeRaw
1# Changelog
2
3## 1.0.71
4
5- Background commands: (Ctrl-b) to run any Bash command in the background so Claude can keep working (great for dev servers, tailing logs, etc.)
6- Customizable status line: add your terminal prompt to Claude Code with /statusline
7
8## 1.0.70
9
10- Performance: Optimized message rendering for better performance with large contexts
11- Windows: Fixed native file search, ripgrep, and subagent functionality
12- Added support for @-mentions in slash command arguments
13
14## 1.0.69
15
16- Upgraded Opus to version 4.1
17
18## 1.0.68
19
20- Fix incorrect model names being used for certain commands like `/pr-comments`
21- Windows: improve permissions checks for allow / deny tools and project trust. This may create a new project entry in `.claude.json` - manually merge the history field if desired.
22- Windows: improve sub-process spawning to eliminate "No such file or directory" when running commands like pnpm
23- Enhanced /doctor command with CLAUDE.md and MCP tool context for self-serve debugging
24- SDK: Added canUseTool callback support for tool confirmation
25- Added `disableAllHooks` setting
26- Improved file suggestions performance in large repos
27
28## 1.0.65
29
30- IDE: Fixed connection stability issues and error handling for diagnostics
31- Windows: Fixed shell environment setup for users without .bashrc files
32
33## 1.0.64
34
35- Agents: Added model customization support - you can now specify which model an agent should use
36- Agents: Fixed unintended access to the recursive agent tool
37- Hooks: Added systemMessage field to hook JSON output for displaying warnings and context
38- SDK: Fixed user input tracking across multi-turn conversations
39- Added hidden files to file search and @-mention suggestions
40
41## 1.0.63
42
43- Windows: Fixed file search, @agent mentions, and custom slash commands functionality
44
45## 1.0.62
46
47- Added @-mention support with typeahead for custom agents. @<your-custom-agent> to invoke it
48- Hooks: Added SessionStart hook for new session initialization
49- /add-dir command now supports typeahead for directory paths
50- Improved network connectivity check reliability
51
52## 1.0.61
53
54- Transcript mode (Ctrl+R): Changed Esc to exit transcript mode rather than interrupt
55- Settings: Added `--settings` flag to load settings from a JSON file
56- Settings: Fixed resolution of settings files paths that are symlinks
57- OTEL: Fixed reporting of wrong organization after authentication changes
58- Slash commands: Fixed permissions checking for allowed-tools with Bash
59- IDE: Added support for pasting images in VSCode MacOS using ⌘+V
60- IDE: Added `CLAUDE_CODE_AUTO_CONNECT_IDE=false` for disabling IDE auto-connection
61- Added `CLAUDE_CODE_SHELL_PREFIX` for wrapping Claude and user-provided shell commands run by Claude Code
62
63## 1.0.60
64
65- You can now create custom subagents for specialized tasks! Run /agents to get started
66
67## 1.0.59
68
69- SDK: Added tool confirmation support with canUseTool callback
70- SDK: Allow specifying env for spawned process
71- Hooks: Exposed PermissionDecision to hooks (including "ask")
72- Hooks: UserPromptSubmit now supports additionalContext in advanced JSON output
73- Fixed issue where some Max users that specified Opus would still see fallback to Sonnet
74
75## 1.0.58
76
77- Added support for reading PDFs
78- MCP: Improved server health status display in 'claude mcp list'
79- Hooks: Added CLAUDE_PROJECT_DIR env var for hook commands
80
81## 1.0.57
82
83- Added support for specifying a model in slash commands
84- Improved permission messages to help Claude understand allowed tools
85- Fix: Remove trailing newlines from bash output in terminal wrapping
86
87## 1.0.56
88
89- Windows: Enabled shift+tab for mode switching on versions of Node.js that support terminal VT mode
90- Fixes for WSL IDE detection
91- Fix an issue causing awsRefreshHelper changes to .aws directory not to be picked up
92
93## 1.0.55
94
95- Clarified knowledge cutoff for Opus 4 and Sonnet 4 models
96- Windows: fixed Ctrl+Z crash
97- SDK: Added ability to capture error logging
98- Add --system-prompt-file option to override system prompt in print mode
99
100## 1.0.54
101
102- Hooks: Added UserPromptSubmit hook and the current working directory to hook inputs
103- Custom slash commands: Added argument-hint to frontmatter
104- Windows: OAuth uses port 45454 and properly constructs browser URL
105- Windows: mode switching now uses alt + m, and plan mode renders properly
106- Shell: Switch to in-memory shell snapshot to fix file-related errors
107
108## 1.0.53
109
110- Updated @-mention file truncation from 100 lines to 2000 lines
111- Add helper script settings for AWS token refresh: awsAuthRefresh (for foreground operations like aws sso login) and awsCredentialExport (for background operation with STS-like response).
112
113## 1.0.52
114
115- Added support for MCP server instructions
116
117## 1.0.51
118
119- Added support for native Windows (requires Git for Windows)
120- Added support for Bedrock API keys through environment variable AWS_BEARER_TOKEN_BEDROCK
121- Settings: /doctor can now help you identify and fix invalid setting files
122- `--append-system-prompt` can now be used in interactive mode, not just --print/-p.
123- Increased auto-compact warning threshold from 60% to 80%
124- Fixed an issue with handling user directories with spaces for shell snapshots
125- OTEL resource now includes os.type, os.version, host.arch, and wsl.version (if running on Windows Subsystem for Linux)
126- Custom slash commands: Fixed user-level commands in subdirectories
127- Plan mode: Fixed issue where rejected plan from sub-task would get discarded
128
129## 1.0.48
130
131- Fixed a bug in v1.0.45 where the app would sometimes freeze on launch
132- Added progress messages to Bash tool based on the last 5 lines of command output
133- Added expanding variables support for MCP server configuration
134- Moved shell snapshots from /tmp to ~/.claude for more reliable Bash tool calls
135- Improved IDE extension path handling when Claude Code runs in WSL
136- Hooks: Added a PreCompact hook
137- Vim mode: Added c, f/F, t/T
138
139## 1.0.45
140
141- Redesigned Search (Grep) tool with new tool input parameters and features
142- Disabled IDE diffs for notebook files, fixing "Timeout waiting after 1000ms" error
143- Fixed config file corruption issue by enforcing atomic writes
144- Updated prompt input undo to Ctrl+\_ to avoid breaking existing Ctrl+U behavior, matching zsh's undo shortcut
145- Stop Hooks: Fixed transcript path after /clear and fixed triggering when loop ends with tool call
146- Custom slash commands: Restored namespacing in command names based on subdirectories. For example, .claude/commands/frontend/component.md is now /frontend:component, not /component.
147
148## 1.0.44
149
150- New /export command lets you quickly export a conversation for sharing
151- MCP: resource_link tool results are now supported
152- MCP: tool annotations and tool titles now display in /mcp view
153- Changed Ctrl+Z to suspend Claude Code. Resume by running `fg`. Prompt input undo is now Ctrl+U.
154
155## 1.0.43
156
157- Fixed a bug where the theme selector was saving excessively
158- Hooks: Added EPIPE system error handling
159
160## 1.0.42
161
162- Added tilde (`~`) expansion support to `/add-dir` command
163
164## 1.0.41
165
166- Hooks: Split Stop hook triggering into Stop and SubagentStop
167- Hooks: Enabled optional timeout configuration for each command
168- Hooks: Added "hook_event_name" to hook input
169- Fixed a bug where MCP tools would display twice in tool list
170- New tool parameters JSON for Bash tool in `tool_decision` event
171
172## 1.0.40
173
174- Fixed a bug causing API connection errors with UNABLE_TO_GET_ISSUER_CERT_LOCALLY if `NODE_EXTRA_CA_CERTS` was set
175
176## 1.0.39
177
178- New Active Time metric in OpenTelemetry logging
179
180## 1.0.38
181
182- Released hooks. Special thanks to community input in https://github.com/anthropics/claude-code/issues/712. Docs: https://docs.anthropic.com/en/docs/claude-code/hooks
183
184## 1.0.37
185
186- Remove ability to set `Proxy-Authorization` header via ANTHROPIC_AUTH_TOKEN or apiKeyHelper
187
188## 1.0.36
189
190- Web search now takes today's date into context
191- Fixed a bug where stdio MCP servers were not terminating properly on exit
192
193## 1.0.35
194
195- Added support for MCP OAuth Authorization Server discovery
196
197## 1.0.34
198
199- Fixed a memory leak causing a MaxListenersExceededWarning message to appear
200
201## 1.0.33
202
203- Improved logging functionality with session ID support
204- Added prompt input undo functionality (Ctrl+Z and vim 'u' command)
205- Improvements to plan mode
206
207## 1.0.32
208
209- Updated loopback config for litellm
210- Added forceLoginMethod setting to bypass login selection screen
211
212## 1.0.31
213
214- Fixed a bug where ~/.claude.json would get reset when file contained invalid JSON
215
216## 1.0.30
217
218- Custom slash commands: Run bash output, @-mention files, enable thinking with thinking keywords
219- Improved file path autocomplete with filename matching
220- Added timestamps in Ctrl-r mode and fixed Ctrl-c handling
221- Enhanced jq regex support for complex filters with pipes and select
222
223## 1.0.29
224
225- Improved CJK character support in cursor navigation and rendering
226
227## 1.0.28
228
229- Slash commands: Fix selector display during history navigation
230- Resizes images before upload to prevent API size limit errors
231- Added XDG_CONFIG_HOME support to configuration directory
232- Performance optimizations for memory usage
233- New attributes (terminal.type, language) in OpenTelemetry logging
234
235## 1.0.27
236
237- Streamable HTTP MCP servers are now supported
238- Remote MCP servers (SSE and HTTP) now support OAuth
239- MCP resources can now be @-mentioned
240- /resume slash command to switch conversations within Claude Code
241
242## 1.0.25
243
244- Slash commands: moved "project" and "user" prefixes to descriptions
245- Slash commands: improved reliability for command discovery
246- Improved support for Ghostty
247- Improved web search reliability
248
249## 1.0.24
250
251- Improved /mcp output
252- Fixed a bug where settings arrays got overwritten instead of merged
253
254## 1.0.23
255
256- Released TypeScript SDK: import @anthropic-ai/claude-code to get started
257- Released Python SDK: pip install claude-code-sdk to get started
258
259## 1.0.22
260
261- SDK: Renamed `total_cost` to `total_cost_usd`
262
263## 1.0.21
264
265- Improved editing of files with tab-based indentation
266- Fix for tool_use without matching tool_result errors
267- Fixed a bug where stdio MCP server processes would linger after quitting Claude Code
268
269## 1.0.18
270
271- Added --add-dir CLI argument for specifying additional working directories
272- Added streaming input support without require -p flag
273- Improved startup performance and session storage performance
274- Added CLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR environment variable to freeze working directory for bash commands
275- Added detailed MCP server tools display (/mcp)
276- MCP authentication and permission improvements
277- Added auto-reconnection for MCP SSE connections on disconnect
278- Fixed issue where pasted content was lost when dialogs appeared
279
280## 1.0.17
281
282- We now emit messages from sub-tasks in -p mode (look for the parent_tool_use_id property)
283- Fixed crashes when the VS Code diff tool is invoked multiple times quickly
284- MCP server list UI improvements
285- Update Claude Code process title to display "claude" instead of "node"
286
287## 1.0.11
288
289- Claude Code can now also be used with a Claude Pro subscription
290- Added /upgrade for smoother switching to Claude Max plans
291- Improved UI for authentication from API keys and Bedrock/Vertex/external auth tokens
292- Improved shell configuration error handling
293- Improved todo list handling during compaction
294
295## 1.0.10
296
297- Added markdown table support
298- Improved streaming performance
299
300## 1.0.8
301
302- Fixed Vertex AI region fallback when using CLOUD_ML_REGION
303- Increased default otel interval from 1s -> 5s
304- Fixed edge cases where MCP_TIMEOUT and MCP_TOOL_TIMEOUT weren't being respected
305- Fixed a regression where search tools unnecessarily asked for permissions
306- Added support for triggering thinking non-English languages
307- Improved compacting UI
308
309## 1.0.7
310
311- Renamed /allowed-tools -> /permissions
312- Migrated allowedTools and ignorePatterns from .claude.json -> settings.json
313- Deprecated claude config commands in favor of editing settings.json
314- Fixed a bug where --dangerously-skip-permissions sometimes didn't work in --print mode
315- Improved error handling for /install-github-app
316- Bugfixes, UI polish, and tool reliability improvements
317
318## 1.0.6
319
320- Improved edit reliability for tab-indented files
321- Respect CLAUDE_CONFIG_DIR everywhere
322- Reduced unnecessary tool permission prompts
323- Added support for symlinks in @file typeahead
324- Bugfixes, UI polish, and tool reliability improvements
325
326## 1.0.4
327
328- Fixed a bug where MCP tool errors weren't being parsed correctly
329
330## 1.0.1
331
332- Added `DISABLE_INTERLEAVED_THINKING` to give users the option to opt out of interleaved thinking.
333- Improved model references to show provider-specific names (Sonnet 3.7 for Bedrock, Sonnet 4 for Console)
334- Updated documentation links and OAuth process descriptions
335
336## 1.0.0
337
338- Claude Code is now generally available
339- Introducing Sonnet 4 and Opus 4 models
340
341## 0.2.125
342
343- Breaking change: Bedrock ARN passed to `ANTHROPIC_MODEL` or `ANTHROPIC_SMALL_FAST_MODEL` should no longer contain an escaped slash (specify `/` instead of `%2F`)
344- Removed `DEBUG=true` in favor of `ANTHROPIC_LOG=debug`, to log all requests
345
346## 0.2.117
347
348- Breaking change: --print JSON output now returns nested message objects, for forwards-compatibility as we introduce new metadata fields
349- Introduced settings.cleanupPeriodDays
350- Introduced CLAUDE_CODE_API_KEY_HELPER_TTL_MS env var
351- Introduced --debug mode
352
353## 0.2.108
354
355- You can now send messages to Claude while it works to steer Claude in real-time
356- Introduced BASH_DEFAULT_TIMEOUT_MS and BASH_MAX_TIMEOUT_MS env vars
357- Fixed a bug where thinking was not working in -p mode
358- Fixed a regression in /cost reporting
359- Deprecated MCP wizard interface in favor of other MCP commands
360- Lots of other bugfixes and improvements
361
362## 0.2.107
363
364- CLAUDE.md files can now import other files. Add @path/to/file.md to ./CLAUDE.md to load additional files on launch
365
366## 0.2.106
367
368- MCP SSE server configs can now specify custom headers
369- Fixed a bug where MCP permission prompt didn't always show correctly
370
371## 0.2.105
372
373- Claude can now search the web
374- Moved system & account status to /status
375- Added word movement keybindings for Vim
376- Improved latency for startup, todo tool, and file edits
377
378## 0.2.102
379
380- Improved thinking triggering reliability
381- Improved @mention reliability for images and folders
382- You can now paste multiple large chunks into one prompt
383
384## 0.2.100
385
386- Fixed a crash caused by a stack overflow error
387- Made db storage optional; missing db support disables --continue and --resume
388
389## 0.2.98
390
391- Fixed an issue where auto-compact was running twice
392
393## 0.2.96
394
395- Claude Code can now also be used with a Claude Max subscription (https://claude.ai/upgrade)
396
397## 0.2.93
398
399- Resume conversations from where you left off from with "claude --continue" and "claude --resume"
400- Claude now has access to a Todo list that helps it stay on track and be more organized
401
402## 0.2.82
403
404- Added support for --disallowedTools
405- Renamed tools for consistency: LSTool -> LS, View -> Read, etc.
406
407## 0.2.75
408
409- Hit Enter to queue up additional messages while Claude is working
410- Drag in or copy/paste image files directly into the prompt
411- @-mention files to directly add them to context
412- Run one-off MCP servers with `claude --mcp-config <path-to-file>`
413- Improved performance for filename auto-complete
414
415## 0.2.74
416
417- Added support for refreshing dynamically generated API keys (via apiKeyHelper), with a 5 minute TTL
418- Task tool can now perform writes and run bash commands
419
420## 0.2.72
421
422- Updated spinner to indicate tokens loaded and tool usage
423
424## 0.2.70
425
426- Network commands like curl are now available for Claude to use
427- Claude can now run multiple web queries in parallel
428- Pressing ESC once immediately interrupts Claude in Auto-accept mode
429
430## 0.2.69
431
432- Fixed UI glitches with improved Select component behavior
433- Enhanced terminal output display with better text truncation logic
434
435## 0.2.67
436
437- Shared project permission rules can be saved in .claude/settings.json
438
439## 0.2.66
440
441- Print mode (-p) now supports streaming output via --output-format=stream-json
442- Fixed issue where pasting could trigger memory or bash mode unexpectedly
443
444## 0.2.63
445
446- Fixed an issue where MCP tools were loaded twice, which caused tool call errors
447
448## 0.2.61
449
450- Navigate menus with vim-style keys (j/k) or bash/emacs shortcuts (Ctrl+n/p) for faster interaction
451- Enhanced image detection for more reliable clipboard paste functionality
452- Fixed an issue where ESC key could crash the conversation history selector
453
454## 0.2.59
455
456- Copy+paste images directly into your prompt
457- Improved progress indicators for bash and fetch tools
458- Bugfixes for non-interactive mode (-p)
459
460## 0.2.54
461
462- Quickly add to Memory by starting your message with '#'
463- Press ctrl+r to see full output for long tool results
464- Added support for MCP SSE transport
465
466## 0.2.53
467
468- New web fetch tool lets Claude view URLs that you paste in
469- Fixed a bug with JPEG detection
470
471## 0.2.50
472
473- New MCP "project" scope now allows you to add MCP servers to .mcp.json files and commit them to your repository
474
475## 0.2.49
476
477- Previous MCP server scopes have been renamed: previous "project" scope is now "local" and "global" scope is now "user"
478
479## 0.2.47
480
481- Press Tab to auto-complete file and folder names
482- Press Shift + Tab to toggle auto-accept for file edits
483- Automatic conversation compaction for infinite conversation length (toggle with /config)
484
485## 0.2.44
486
487- Ask Claude to make a plan with thinking mode: just say 'think' or 'think harder' or even 'ultrathink'
488
489## 0.2.41
490
491- MCP server startup timeout can now be configured via MCP_TIMEOUT environment variable
492- MCP server startup no longer blocks the app from starting up
493
494## 0.2.37
495
496- New /release-notes command lets you view release notes at any time
497- `claude config add/remove` commands now accept multiple values separated by commas or spaces
498
499## 0.2.36
500
501- Import MCP servers from Claude Desktop with `claude mcp add-from-claude-desktop`
502- Add MCP servers as JSON strings with `claude mcp add-json <n> <json>`
503
504## 0.2.34
505
506- Vim bindings for text input - enable with /vim or /config
507
508## 0.2.32
509
510- Interactive MCP setup wizard: Run "claude mcp add" to add MCP servers with a step-by-step interface
511- Fix for some PersistentShell issues
512
513## 0.2.31
514
515- Custom slash commands: Markdown files in .claude/commands/ directories now appear as custom slash commands to insert prompts into your conversation
516- MCP debug mode: Run with --mcp-debug flag to get more information about MCP server errors
517
518## 0.2.30
519
520- Added ANSI color theme for better terminal compatibility
521- Fixed issue where slash command arguments weren't being sent properly
522- (Mac-only) API keys are now stored in macOS Keychain
523
524## 0.2.26
525
526- New /approved-tools command for managing tool permissions
527- Word-level diff display for improved code readability
528- Fuzzy matching for slash commands
529
530## 0.2.21
531
532- Fuzzy matching for /commands