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 3fd750e · on Aug 7, 2025 History
531 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
7## 1.0.70
8
9- Performance: Optimized message rendering for better performance with large contexts
10- Windows: Fixed native file search, ripgrep, and subagent functionality
11- Added support for @-mentions in slash command arguments
12
13## 1.0.69
14
15- Upgraded Opus to version 4.1
16
17## 1.0.68
18
19- Fix incorrect model names being used for certain commands like `/pr-comments`
20- 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.
21- Windows: improve sub-process spawning to eliminate "No such file or directory" when running commands like pnpm
22- Enhanced /doctor command with CLAUDE.md and MCP tool context for self-serve debugging
23- SDK: Added canUseTool callback support for tool confirmation
24- Added `disableAllHooks` setting
25- Improved file suggestions performance in large repos
26
27## 1.0.65
28
29- IDE: Fixed connection stability issues and error handling for diagnostics
30- Windows: Fixed shell environment setup for users without .bashrc files
31
32## 1.0.64
33
34- Agents: Added model customization support - you can now specify which model an agent should use
35- Agents: Fixed unintended access to the recursive agent tool
36- Hooks: Added systemMessage field to hook JSON output for displaying warnings and context
37- SDK: Fixed user input tracking across multi-turn conversations
38- Added hidden files to file search and @-mention suggestions
39
40## 1.0.63
41
42- Windows: Fixed file search, @agent mentions, and custom slash commands functionality
43
44## 1.0.62
45
46- Added @-mention support with typeahead for custom agents. @<your-custom-agent> to invoke it
47- Hooks: Added SessionStart hook for new session initialization
48- /add-dir command now supports typeahead for directory paths
49- Improved network connectivity check reliability
50
51## 1.0.61
52
53- Transcript mode (Ctrl+R): Changed Esc to exit transcript mode rather than interrupt
54- Settings: Added `--settings` flag to load settings from a JSON file
55- Settings: Fixed resolution of settings files paths that are symlinks
56- OTEL: Fixed reporting of wrong organization after authentication changes
57- Slash commands: Fixed permissions checking for allowed-tools with Bash
58- IDE: Added support for pasting images in VSCode MacOS using ⌘+V
59- IDE: Added `CLAUDE_CODE_AUTO_CONNECT_IDE=false` for disabling IDE auto-connection
60- Added `CLAUDE_CODE_SHELL_PREFIX` for wrapping Claude and user-provided shell commands run by Claude Code
61
62## 1.0.60
63
64- You can now create custom subagents for specialized tasks! Run /agents to get started
65
66## 1.0.59
67
68- SDK: Added tool confirmation support with canUseTool callback
69- SDK: Allow specifying env for spawned process
70- Hooks: Exposed PermissionDecision to hooks (including "ask")
71- Hooks: UserPromptSubmit now supports additionalContext in advanced JSON output
72- Fixed issue where some Max users that specified Opus would still see fallback to Sonnet
73
74## 1.0.58
75
76- Added support for reading PDFs
77- MCP: Improved server health status display in 'claude mcp list'
78- Hooks: Added CLAUDE_PROJECT_DIR env var for hook commands
79
80## 1.0.57
81
82- Added support for specifying a model in slash commands
83- Improved permission messages to help Claude understand allowed tools
84- Fix: Remove trailing newlines from bash output in terminal wrapping
85
86## 1.0.56
87
88- Windows: Enabled shift+tab for mode switching on versions of Node.js that support terminal VT mode
89- Fixes for WSL IDE detection
90- Fix an issue causing awsRefreshHelper changes to .aws directory not to be picked up
91
92## 1.0.55
93
94- Clarified knowledge cutoff for Opus 4 and Sonnet 4 models
95- Windows: fixed Ctrl+Z crash
96- SDK: Added ability to capture error logging
97- Add --system-prompt-file option to override system prompt in print mode
98
99## 1.0.54
100
101- Hooks: Added UserPromptSubmit hook and the current working directory to hook inputs
102- Custom slash commands: Added argument-hint to frontmatter
103- Windows: OAuth uses port 45454 and properly constructs browser URL
104- Windows: mode switching now uses alt + m, and plan mode renders properly
105- Shell: Switch to in-memory shell snapshot to fix file-related errors
106
107## 1.0.53
108
109- Updated @-mention file truncation from 100 lines to 2000 lines
110- 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).
111
112## 1.0.52
113
114- Added support for MCP server instructions
115
116## 1.0.51
117
118- Added support for native Windows (requires Git for Windows)
119- Added support for Bedrock API keys through environment variable AWS_BEARER_TOKEN_BEDROCK
120- Settings: /doctor can now help you identify and fix invalid setting files
121- `--append-system-prompt` can now be used in interactive mode, not just --print/-p.
122- Increased auto-compact warning threshold from 60% to 80%
123- Fixed an issue with handling user directories with spaces for shell snapshots
124- OTEL resource now includes os.type, os.version, host.arch, and wsl.version (if running on Windows Subsystem for Linux)
125- Custom slash commands: Fixed user-level commands in subdirectories
126- Plan mode: Fixed issue where rejected plan from sub-task would get discarded
127
128## 1.0.48
129
130- Fixed a bug in v1.0.45 where the app would sometimes freeze on launch
131- Added progress messages to Bash tool based on the last 5 lines of command output
132- Added expanding variables support for MCP server configuration
133- Moved shell snapshots from /tmp to ~/.claude for more reliable Bash tool calls
134- Improved IDE extension path handling when Claude Code runs in WSL
135- Hooks: Added a PreCompact hook
136- Vim mode: Added c, f/F, t/T
137
138## 1.0.45
139
140- Redesigned Search (Grep) tool with new tool input parameters and features
141- Disabled IDE diffs for notebook files, fixing "Timeout waiting after 1000ms" error
142- Fixed config file corruption issue by enforcing atomic writes
143- Updated prompt input undo to Ctrl+\_ to avoid breaking existing Ctrl+U behavior, matching zsh's undo shortcut
144- Stop Hooks: Fixed transcript path after /clear and fixed triggering when loop ends with tool call
145- Custom slash commands: Restored namespacing in command names based on subdirectories. For example, .claude/commands/frontend/component.md is now /frontend:component, not /component.
146
147## 1.0.44
148
149- New /export command lets you quickly export a conversation for sharing
150- MCP: resource_link tool results are now supported
151- MCP: tool annotations and tool titles now display in /mcp view
152- Changed Ctrl+Z to suspend Claude Code. Resume by running `fg`. Prompt input undo is now Ctrl+U.
153
154## 1.0.43
155
156- Fixed a bug where the theme selector was saving excessively
157- Hooks: Added EPIPE system error handling
158
159## 1.0.42
160
161- Added tilde (`~`) expansion support to `/add-dir` command
162
163## 1.0.41
164
165- Hooks: Split Stop hook triggering into Stop and SubagentStop
166- Hooks: Enabled optional timeout configuration for each command
167- Hooks: Added "hook_event_name" to hook input
168- Fixed a bug where MCP tools would display twice in tool list
169- New tool parameters JSON for Bash tool in `tool_decision` event
170
171## 1.0.40
172
173- Fixed a bug causing API connection errors with UNABLE_TO_GET_ISSUER_CERT_LOCALLY if `NODE_EXTRA_CA_CERTS` was set
174
175## 1.0.39
176
177- New Active Time metric in OpenTelemetry logging
178
179## 1.0.38
180
181- 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
182
183## 1.0.37
184
185- Remove ability to set `Proxy-Authorization` header via ANTHROPIC_AUTH_TOKEN or apiKeyHelper
186
187## 1.0.36
188
189- Web search now takes today's date into context
190- Fixed a bug where stdio MCP servers were not terminating properly on exit
191
192## 1.0.35
193
194- Added support for MCP OAuth Authorization Server discovery
195
196## 1.0.34
197
198- Fixed a memory leak causing a MaxListenersExceededWarning message to appear
199
200## 1.0.33
201
202- Improved logging functionality with session ID support
203- Added prompt input undo functionality (Ctrl+Z and vim 'u' command)
204- Improvements to plan mode
205
206## 1.0.32
207
208- Updated loopback config for litellm
209- Added forceLoginMethod setting to bypass login selection screen
210
211## 1.0.31
212
213- Fixed a bug where ~/.claude.json would get reset when file contained invalid JSON
214
215## 1.0.30
216
217- Custom slash commands: Run bash output, @-mention files, enable thinking with thinking keywords
218- Improved file path autocomplete with filename matching
219- Added timestamps in Ctrl-r mode and fixed Ctrl-c handling
220- Enhanced jq regex support for complex filters with pipes and select
221
222## 1.0.29
223
224- Improved CJK character support in cursor navigation and rendering
225
226## 1.0.28
227
228- Slash commands: Fix selector display during history navigation
229- Resizes images before upload to prevent API size limit errors
230- Added XDG_CONFIG_HOME support to configuration directory
231- Performance optimizations for memory usage
232- New attributes (terminal.type, language) in OpenTelemetry logging
233
234## 1.0.27
235
236- Streamable HTTP MCP servers are now supported
237- Remote MCP servers (SSE and HTTP) now support OAuth
238- MCP resources can now be @-mentioned
239- /resume slash command to switch conversations within Claude Code
240
241## 1.0.25
242
243- Slash commands: moved "project" and "user" prefixes to descriptions
244- Slash commands: improved reliability for command discovery
245- Improved support for Ghostty
246- Improved web search reliability
247
248## 1.0.24
249
250- Improved /mcp output
251- Fixed a bug where settings arrays got overwritten instead of merged
252
253## 1.0.23
254
255- Released TypeScript SDK: import @anthropic-ai/claude-code to get started
256- Released Python SDK: pip install claude-code-sdk to get started
257
258## 1.0.22
259
260- SDK: Renamed `total_cost` to `total_cost_usd`
261
262## 1.0.21
263
264- Improved editing of files with tab-based indentation
265- Fix for tool_use without matching tool_result errors
266- Fixed a bug where stdio MCP server processes would linger after quitting Claude Code
267
268## 1.0.18
269
270- Added --add-dir CLI argument for specifying additional working directories
271- Added streaming input support without require -p flag
272- Improved startup performance and session storage performance
273- Added CLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR environment variable to freeze working directory for bash commands
274- Added detailed MCP server tools display (/mcp)
275- MCP authentication and permission improvements
276- Added auto-reconnection for MCP SSE connections on disconnect
277- Fixed issue where pasted content was lost when dialogs appeared
278
279## 1.0.17
280
281- We now emit messages from sub-tasks in -p mode (look for the parent_tool_use_id property)
282- Fixed crashes when the VS Code diff tool is invoked multiple times quickly
283- MCP server list UI improvements
284- Update Claude Code process title to display "claude" instead of "node"
285
286## 1.0.11
287
288- Claude Code can now also be used with a Claude Pro subscription
289- Added /upgrade for smoother switching to Claude Max plans
290- Improved UI for authentication from API keys and Bedrock/Vertex/external auth tokens
291- Improved shell configuration error handling
292- Improved todo list handling during compaction
293
294## 1.0.10
295
296- Added markdown table support
297- Improved streaming performance
298
299## 1.0.8
300
301- Fixed Vertex AI region fallback when using CLOUD_ML_REGION
302- Increased default otel interval from 1s -> 5s
303- Fixed edge cases where MCP_TIMEOUT and MCP_TOOL_TIMEOUT weren't being respected
304- Fixed a regression where search tools unnecessarily asked for permissions
305- Added support for triggering thinking non-English languages
306- Improved compacting UI
307
308## 1.0.7
309
310- Renamed /allowed-tools -> /permissions
311- Migrated allowedTools and ignorePatterns from .claude.json -> settings.json
312- Deprecated claude config commands in favor of editing settings.json
313- Fixed a bug where --dangerously-skip-permissions sometimes didn't work in --print mode
314- Improved error handling for /install-github-app
315- Bugfixes, UI polish, and tool reliability improvements
316
317## 1.0.6
318
319- Improved edit reliability for tab-indented files
320- Respect CLAUDE_CONFIG_DIR everywhere
321- Reduced unnecessary tool permission prompts
322- Added support for symlinks in @file typeahead
323- Bugfixes, UI polish, and tool reliability improvements
324
325## 1.0.4
326
327- Fixed a bug where MCP tool errors weren't being parsed correctly
328
329## 1.0.1
330
331- Added `DISABLE_INTERLEAVED_THINKING` to give users the option to opt out of interleaved thinking.
332- Improved model references to show provider-specific names (Sonnet 3.7 for Bedrock, Sonnet 4 for Console)
333- Updated documentation links and OAuth process descriptions
334
335## 1.0.0
336
337- Claude Code is now generally available
338- Introducing Sonnet 4 and Opus 4 models
339
340## 0.2.125
341
342- Breaking change: Bedrock ARN passed to `ANTHROPIC_MODEL` or `ANTHROPIC_SMALL_FAST_MODEL` should no longer contain an escaped slash (specify `/` instead of `%2F`)
343- Removed `DEBUG=true` in favor of `ANTHROPIC_LOG=debug`, to log all requests
344
345## 0.2.117
346
347- Breaking change: --print JSON output now returns nested message objects, for forwards-compatibility as we introduce new metadata fields
348- Introduced settings.cleanupPeriodDays
349- Introduced CLAUDE_CODE_API_KEY_HELPER_TTL_MS env var
350- Introduced --debug mode
351
352## 0.2.108
353
354- You can now send messages to Claude while it works to steer Claude in real-time
355- Introduced BASH_DEFAULT_TIMEOUT_MS and BASH_MAX_TIMEOUT_MS env vars
356- Fixed a bug where thinking was not working in -p mode
357- Fixed a regression in /cost reporting
358- Deprecated MCP wizard interface in favor of other MCP commands
359- Lots of other bugfixes and improvements
360
361## 0.2.107
362
363- CLAUDE.md files can now import other files. Add @path/to/file.md to ./CLAUDE.md to load additional files on launch
364
365## 0.2.106
366
367- MCP SSE server configs can now specify custom headers
368- Fixed a bug where MCP permission prompt didn't always show correctly
369
370## 0.2.105
371
372- Claude can now search the web
373- Moved system & account status to /status
374- Added word movement keybindings for Vim
375- Improved latency for startup, todo tool, and file edits
376
377## 0.2.102
378
379- Improved thinking triggering reliability
380- Improved @mention reliability for images and folders
381- You can now paste multiple large chunks into one prompt
382
383## 0.2.100
384
385- Fixed a crash caused by a stack overflow error
386- Made db storage optional; missing db support disables --continue and --resume
387
388## 0.2.98
389
390- Fixed an issue where auto-compact was running twice
391
392## 0.2.96
393
394- Claude Code can now also be used with a Claude Max subscription (https://claude.ai/upgrade)
395
396## 0.2.93
397
398- Resume conversations from where you left off from with "claude --continue" and "claude --resume"
399- Claude now has access to a Todo list that helps it stay on track and be more organized
400
401## 0.2.82
402
403- Added support for --disallowedTools
404- Renamed tools for consistency: LSTool -> LS, View -> Read, etc.
405
406## 0.2.75
407
408- Hit Enter to queue up additional messages while Claude is working
409- Drag in or copy/paste image files directly into the prompt
410- @-mention files to directly add them to context
411- Run one-off MCP servers with `claude --mcp-config <path-to-file>`
412- Improved performance for filename auto-complete
413
414## 0.2.74
415
416- Added support for refreshing dynamically generated API keys (via apiKeyHelper), with a 5 minute TTL
417- Task tool can now perform writes and run bash commands
418
419## 0.2.72
420
421- Updated spinner to indicate tokens loaded and tool usage
422
423## 0.2.70
424
425- Network commands like curl are now available for Claude to use
426- Claude can now run multiple web queries in parallel
427- Pressing ESC once immediately interrupts Claude in Auto-accept mode
428
429## 0.2.69
430
431- Fixed UI glitches with improved Select component behavior
432- Enhanced terminal output display with better text truncation logic
433
434## 0.2.67
435
436- Shared project permission rules can be saved in .claude/settings.json
437
438## 0.2.66
439
440- Print mode (-p) now supports streaming output via --output-format=stream-json
441- Fixed issue where pasting could trigger memory or bash mode unexpectedly
442
443## 0.2.63
444
445- Fixed an issue where MCP tools were loaded twice, which caused tool call errors
446
447## 0.2.61
448
449- Navigate menus with vim-style keys (j/k) or bash/emacs shortcuts (Ctrl+n/p) for faster interaction
450- Enhanced image detection for more reliable clipboard paste functionality
451- Fixed an issue where ESC key could crash the conversation history selector
452
453## 0.2.59
454
455- Copy+paste images directly into your prompt
456- Improved progress indicators for bash and fetch tools
457- Bugfixes for non-interactive mode (-p)
458
459## 0.2.54
460
461- Quickly add to Memory by starting your message with '#'
462- Press ctrl+r to see full output for long tool results
463- Added support for MCP SSE transport
464
465## 0.2.53
466
467- New web fetch tool lets Claude view URLs that you paste in
468- Fixed a bug with JPEG detection
469
470## 0.2.50
471
472- New MCP "project" scope now allows you to add MCP servers to .mcp.json files and commit them to your repository
473
474## 0.2.49
475
476- Previous MCP server scopes have been renamed: previous "project" scope is now "local" and "global" scope is now "user"
477
478## 0.2.47
479
480- Press Tab to auto-complete file and folder names
481- Press Shift + Tab to toggle auto-accept for file edits
482- Automatic conversation compaction for infinite conversation length (toggle with /config)
483
484## 0.2.44
485
486- Ask Claude to make a plan with thinking mode: just say 'think' or 'think harder' or even 'ultrathink'
487
488## 0.2.41
489
490- MCP server startup timeout can now be configured via MCP_TIMEOUT environment variable
491- MCP server startup no longer blocks the app from starting up
492
493## 0.2.37
494
495- New /release-notes command lets you view release notes at any time
496- `claude config add/remove` commands now accept multiple values separated by commas or spaces
497
498## 0.2.36
499
500- Import MCP servers from Claude Desktop with `claude mcp add-from-claude-desktop`
501- Add MCP servers as JSON strings with `claude mcp add-json <n> <json>`
502
503## 0.2.34
504
505- Vim bindings for text input - enable with /vim or /config
506
507## 0.2.32
508
509- Interactive MCP setup wizard: Run "claude mcp add" to add MCP servers with a step-by-step interface
510- Fix for some PersistentShell issues
511
512## 0.2.31
513
514- Custom slash commands: Markdown files in .claude/commands/ directories now appear as custom slash commands to insert prompts into your conversation
515- MCP debug mode: Run with --mcp-debug flag to get more information about MCP server errors
516
517## 0.2.30
518
519- Added ANSI color theme for better terminal compatibility
520- Fixed issue where slash command arguments weren't being sent properly
521- (Mac-only) API keys are now stored in macOS Keychain
522
523## 0.2.26
524
525- New /approved-tools command for managing tool permissions
526- Word-level diff display for improved code readability
527- Fuzzy matching for slash commands
528
529## 0.2.21
530
531- Fuzzy matching for /commands