Claude Code changelog changedchangelog
Nearest release: v2.1.308, published under an hour before upstream edited the page. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Upstream edited this page at 22 Oct 2026 04:47 UTC, give or take a minute or two: the time comes from Anthropic’s own sitemap rather than from a commit. This site recorded the change at 22 Oct 2026 18:07 UTC.
Upstream edited
Recorded here
Lines+873added
Lines−17removed
From line
8
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits72to this page, all time
The whole hunk
from line 8, old and new numbered
/
This page is larger than the 256 KiB this site keeps, so one side of the diff below stops where the stored text does.
from line 8
88
99**2.1.308**
1010
11October 22, 2026
12
1113* Bug fixes and reliability improvements
1214
1315**2.1.307**
1416
17October 21, 2026
18
1519**Added**
1620
1721* Added `/claude-api api-version` to move Claude API code to API version `2026-10-21`
from line 73
6973
7074**2.1.306**
7175
76October 21, 2026
77
7278* Fixed `claude -p` inside cloud sessions reporting "Not logged in" when started without the session's environment variables
7379* Fixed an API error being replaced by "prompt is too long", and an unneeded compaction, in sessions with many MCP tools
7480
7581**2.1.305**
7682
83October 20, 2026
84
7785**Added**
7886
7987* Added `sandbox.allowNestedUserNamespaces` so sandboxed commands on Linux can create user namespaces (Chromium, rootless podman); this lets them get around the sandbox's file protections
from line 134
126134
127135**2.1.304**
128136
137October 19, 2026
138
129139**Added**
130140
131141* Added a transcript line when a plugin that runs first skips one of your mod's hooks, naming the event and the plugin
from line 226
216226
217227**2.1.303**
218228
229October 18, 2026
230
219231* Bug fixes and reliability improvements
220232
221233**2.1.302**
222234
235October 16, 2026
236
223237**Added**
224238
225239* Added `CLAUDE_CODE_AUTO_MODE_MODEL` to choose the model the client-side auto mode classifier uses; it does not change the model for server-side auto mode
from line 311
297311
298312**2.1.301**
299313
314October 15, 2026
315
300316**Added**
301317
302318* Added `CLAUDE_CODE_RETRY_WATCHDOG_MAX_USAGE_LIMIT_WAIT_MS` so unattended retry mode (`CLAUDE_CODE_RETRY_WATCHDOG`) ends a request at a usage limit that resets too far away instead of waiting
from line 378
362378
363379**2.1.300**
364380
381October 14, 2026
382
365383**Added**
366384
367385* Added `virtualText` to a mod's prompt decorations: text drawn in the prompt box that never enters the draft or the submitted prompt
from line 482
464482
465483**2.1.299**
466484
485October 14, 2026
486
467487**Added**
468488
469489* Added `CLAUDE_CODE_DISABLE_FILE_MENTIONS` environment variable: Claude gets the paths of `@` file mentions to read with its file tools, not the files' contents
from line 688
668688
669689**2.1.296**
670690
691October 9, 2026
692
671693**Added**
672694
673695* Added a `code` key to the Claude apps gateway's `managed.policies[]`: the same settings as `cli`, also applied in Claude Desktop's Code tab; beside `desktop`, it turns on Claude Desktop's gateway mode
from line 783
761783
762784**2.1.295**
763785
786October 8, 2026
787
764788**Added**
765789
766790* Added `onFailure: "block"` for command and HTTP hooks: a hook that can't start, times out, or exits with an unexpected code blocks the action instead of letting it through
from line 942
918942
919943**2.1.294**
920944
945October 8, 2026
946
921947* Fixed `prompt` and `agent` hooks written as instructions (such as "Block commands that...") allowing what they should block
922948* Improved how `prompt` hooks on Stop and SubagentStop written as instructions (such as "Carry on if the build is broken") are judged, so Claude is less likely to stop early
923949
924950**2.1.293**
925951
952October 7, 2026
953
926954**Added**
927955
928956* Added Claude Haiku 5.5 (`claude-haiku-5-5`), now the default Haiku model on the Anthropic API — 1M context, $0.10/$0.50 per Mtok ($0.50/$2.50 for prompts over 100K)
from line 1021
9931021
9941022**2.1.292**
9951023
1024October 6, 2026
1025
9961026**Added**
9971027
9981028* Added `--marketplace <source>` to `claude plugin install`: adds the marketplace if needed, under the same policy checks as `claude plugin marketplace add`, then installs the plugin from it
from line 1129
10991129
11001130**2.1.291**
11011131
1132October 6, 2026
1133
11021134* Fixed a regression in 2.1.290 where cloud sessions could drop answers to permission prompts
11031135* Fixed a regression in 2.1.288 where the last messages of a session could be lost when quitting
11041136
11051137**2.1.290**
11061138
1139October 5, 2026
1140
11071141**Added**
11081142
11091143* Added `serverToolUses` to the result of a mod's `turn.step` hook: the tool calls the API ran itself (the advisor), each with its id, name, input, start and end
from line 1342
13081342
13091343**2.1.289**
13101344
1345October 3, 2026
1346
13111347**Added**
13121348
13131349* Added `agent.spawn` for teammates, one agent id across plugin hook events, and idle and waiting states in `$.agent.list()`
from line 1385
13491385
13501386**2.1.288**
13511387
1388October 2, 2026
1389
13521390**Added**
13531391
13541392* Added `$.ui.selection()` for mods: returns the text you last selected in fullscreen mode and, when the selection lies within one transcript row, that row
from line 1490
14521490
14531491**2.1.287**
14541492
1493October 1, 2026
1494
14551495**Added**
14561496
14571497* Added Claude Mods: plugins may now modify deeper behavior
from line 1612
15721612
15731613**2.1.286**
15741614
1615September 30, 2026
1616
15751617**Added**
15761618
15771619* Added a count such as "2 of 5" to the permission prompt when several permission requests stack up
from line 1716
16741716
16751717**2.1.285**
16761718
1719September 29, 2026
1720
16771721**Added**
16781722
16791723* Added `CLAUDE_CODE_DISABLE_WEB_FETCH` environment variable to turn off the WebFetch tool
from line 1868
18241868
18251869**2.1.284**
18261870
1871September 28, 2026
1872
18271873**Added**
18281874
18291875* Added Claude Sonnet 5.5 (`claude-sonnet-5-5`), now the default Sonnet model on the Anthropic API — 1M context, $2/$10 per Mtok with \$0.20/Mtok cache reads
from line 1984
19381984
19391985**2.1.283**
19401986
1987September 25, 2026
1988
19411989**Added**
19421990
19431991* Added `x-claude-code-prompt-id` to the gateway hint headers so LLM gateways can group the requests that serve one user prompt; opt in with `CLAUDE_CODE_GATEWAY_HINT_HEADERS=1`
from line 2066
20182066* Fixed `keybindings.json` silently accepting a misspelled modifier such as `ctl+k`; it now warns in the debug log and suggests the fix
20192067* Fixed footer hints still saying "Enter to view" after `footer:openSelected` was rebound or unbound in `keybindings.json`
20202068* Fixed keys typed quickly together (type-ahead, key repeat, bursts over ssh or tmux) sometimes being handled against stale state
2021* Fixed worktree checkouts failing certificate verification (for example on Git LFS downloads) when the CA certificate is passed to git as `GIT_CONFIG_COUNT` environment pairs
2022* Fixed sandboxed `git` asking credential helpers to store the sandbox proxy's login, which printed "failed to store"
2023* Fixed managed `sandbox` settings being ignored entirely when one nested value was invalid; the invalid value now fails closed and the rest of the block still applies
2024* Fixed Claude's edits to its own auto-memory notes
2069* Fixed worktree checkouts failing certificate verification (for example on Git LFS
No line in this hunk matches that.