Every build, month by month
Every published build with a read, grouped by the month npm dated it. One row each: what came out, how much of it there was, and whether Anthropic ever wrote it up themselves.
1 entry, 0 of them switched off
333 entries, 8 of them switched off
Claude Code can now use the claude.ai sign-in passed over by the desktop app that started it. That lent sign-in works like CLAUDE_CODE_OAUTH_TOKEN, but plugins cannot use it…
2 entries, 0 of them switched off
Claude Code now reads a token from its standard credential file in local sessions too. Before this, it only read that file when CLAUDE_CODE_REMOTE was set, the variable that marks…
310 entries, 39 of them switched off
Claude can now upload a local file to Google Drive by naming it, up to 10 MB. Claude Code reads the file only after you approve the call. In remote sessions that support it, you…
448 entries, 8 of them switched off
Claude Code's bundled default model is now Opus 5.5 at medium effort, replacing Opus 5 at high effort. You can set CLAUDE_CODE_SKIP_PLUGIN_LOCAL_MCP_SERVERS to keep only the…
5 entries, 0 of them switched off
Claude Code now refuses to trust four blocked root certificates, wherever they come from. They are dropped from every certificate source, including a file you point to with…
269 entries, 7 of them switched off
A bundled chart plugin now lets Claude draw charts right in the transcript. It also charts metrics from MCP tools automatically, and the Auto charts switch in /config controls…
321 entries, 13 of them switched off
You can now set CLAUDE_CODE_RETRY_WATCHDOG_MAX_USAGE_LIMIT_WAIT_MS to make a request fail instead of waiting past that time after hitting a usage limit. This release has 272…
408 entries, 6 of them switched off
Sessions can now start in a sandbox VM by default when an option is set. The new --unsandboxed flag opts out of that and of cloud-by-default too. Setting…
529 entries, 28 of them switched off
You can now set subagentInheritVerifiedTurns to true so subagents start with your earlier verified messages. MCP servers can now provide skills to Claude Code. Each skill's size…
370 entries, 11 of them switched off
MCP tools that take file inputs can now accept local file paths. Claude Code uploads the file and passes the tool a reference instead. The Read tool's new allow_large option…
485 entries, 19 of them switched off
Setting CLAUDE_CODE_RESTRICT_PERSONAL_CONFIG now stops your own skills, agents and hooks from widening permissions. In that mode, errors in your user settings file block every…
6 entries, 1 of them switched off
Hooks that ask a model to judge an action now treat its answer as a plain allow or block. An "ok" value of true lets the action go ahead, and false stops it. Prompt hooks now ask…
338 entries, 5 of them switched off
The haiku model name now picks Claude Haiku 5.5 on Anthropic's own API. Bedrock, Vertex and other providers stay on Haiku 4.5. Haiku 5.5 has a 1M-token context window and its own…
392 entries, 17 of them switched off
You can now name backup models in the new accessFallbackModels setting, for use when the main model can't be used. A new per-server disableAutoBackground option stops Claude Code…
10 entries, 1 of them switched off
Session transcripts now load in a single pass again when you resume, without pausing or retrying. Claude Code no longer stops partway to let the interface catch up. It also no…
620 entries, 8 of them switched off
You can now stop Claude Code compacting a long conversation while you sit idle by setting idleCompaction to false. The CLAUDE_CODE_IDLE_COMPACT_MIN_TOKENS variable sets how large…
67 entries, 0 of them switched off
You can now let plugins read the text you have selected in a connected app, in sessions without a terminal. The first non-empty answer that arrives within 5 seconds is used…
425 entries, 27 of them switched off
You can now hold project settings changes made from a cloud session until you approve them, by running claude apply-project-settings in that folder. This applies when your…
526 entries, 9 of them switched off
You can now run gh api REST calls in sessions that do not have the GitHub CLI installed. Claude Code supplies its own limited gh and sends those calls through the session's…
348 entries, 25 of them switched off
You can now rely on a built-in security plugin to keep your deny rules in force when other plugins try to lift them. It can also refuse plugins that did not come from your…
503 entries, 9 of them switched off
You can now use claude-sonnet-5 with high effort and automatic thinking switched on by default. The high effort option is now labelled Recommended instead of Default. Admins can…
434 entries, 40 of them switched off
You can now use Sonnet 5.5 by choosing sonnet on the Anthropic API, where it replaces Sonnet 5. Bedrock, Vertex and other providers still get their older models under that name…
486 entries, 26 of them switched off
You can now have a new bundled skill write live mods that reload inside your session once you agree. A mod is a plugin that adds panes, status lines, toasts or hooks to Claude…
430 entries, 32 of them switched off
You can now set maxProseWidth to wrap Claude's text replies at a fixed column width, while code and tables keep the full terminal width. Claude Code now retries once on the same…
1,240 entries, 65 of them switched off
Of this build's 31 gated additions, none are switched on yet. They include code for a headless claude -p --cloud run whose enabling switch returns false, and an Anthropic plugin…
771 entries, 27 of them switched off
Twelve of this build's 27 gated additions stay out of reach. A warning for dispatching multiple write-capable agents into the same directory sits unreleased alongside a new…
27 entries, 7 of them switched off
This build's seven gated additions cluster around a single unreleased feature: a billing notice warning some users that auto mode is still charging them for classifier requests…
527 entries, 33 of them switched off
This build carries 33 gated additions that stay out of reach. AGENTS.md fallback now becomes the default instruction-file mode when a project has no CLAUDE.md, with the…
2 entries, 0 of them switched off
Model-advisor's standby model selection now checks not just whether a fallback model is available but whether the account is actually talking to Anthropic's own API before picking…
464 entries, 54 of them switched off
This build carries 54 gated additions that stay out of reach. The largest is a new post-turn memory system: a 'memory' settings layer plus an MCP tool for writing, reading…
425 entries, 28 of them switched off
This build carries 28 gated additions that stay out of reach. A pasted-content wrapping tag can mark text as untrusted content, but the killswitch tengu_virtual_pancake defaults…
278 entries, 15 of them switched off
This build carries 15 gated additions, none of them reachable yet. A new OAuth scope, user:plugins, is requested only when PLUGINS_SCOPE_REGISTERED is set, and remote teleport…
18 entries, 10 of them switched off
This build adds a new AppifactRepl tool that runs a full JavaScript program against an Artifact's data in one shot instead of one call at a time, including a claude.see() helper…
567 entries, 35 of them switched off
This build carries 35 gated additions that stay out of reach. A new turn_handoff protocol lets a session hand off an in-progress turn, including one with pending tool calls, to a…
2 entries, 1 of them switched off
This build adds a bypass to the refusal-fallback "silent rearm" system, letting Claude Code re-offer a fallback model for bio-category refusals even after a user has already…
518 entries, 24 of them switched off
This build carries 25 gated additions that stay switched off. A new 'bashEditDiff' system snapshots the repo around Bash commands to track their file changes, controlled by an env…
422 entries, 20 of them switched off
This build carries 26 dormant additions, most left unswitched. A new allowed_domains parameter lets the model tell a sandboxed Bash, PowerShell, or Monitor command which network…
366 entries, 35 of them switched off
This build carries 36 dormant additions, none of them switched on yet. The largest thread is around artifacts: a new asset store for uploading, listing, reading and deleting…
3 entries, 0 of them switched off
The Cloud gateway sign-in error message got trimmed this release, dropping the instructions that used to point to the CLAUDE_CODE_USE_GATEWAY, ANTHROPIC_BASE_URL and…
511 entries, 50 of them switched off
This build carries 50 gated additions with nothing switched on, the largest a new "tether" request-threading protocol that would let the API thread requests together server-side…
14 entries, 8 of them switched off
This build carries eight dormant additions, none of them reachable yet. A hidden environment-variable switch called tengu_polished_dewdrop can make certain responses silently drop…
189 entries, 21 of them switched off
This build carries 17 dormant additions, most switched off behind hardcoded flags. Real-time PR webhook updates exist but are hardcoded off, so PR monitoring still polls every 30…
400 entries, 53 of them switched off
This build carries 42 gated entries, most not switched on. A hidden 'advisor' local command exists to let Claude consult a stronger model, but it stays off unless internal gates…
210 entries, 26 of them switched off
Eighteen entries in this build sit behind flags or unfinished wiring. Plugin hooks can now be written as real JS modules, but they stay off unless…
3 entries, 0 of them switched off
525 entries, 78 of them switched off
Forty-seven items sit in this build without a way to reach them yet. A Mythos 5.1 model has been added to the catalog with a 1M context window and max effort levels, and the…
23 entries, 4 of them switched off
Four features sit in this build without a way to reach them yet. A new "instructionFiles" context channel exists behind the carved-slate flag, part of a static system-prompt…
378 entries, 36 of them switched off
Sixty-four entries sit in this build without a way to reach them. Live presence on artifacts, which tracks who is viewing one over a websocket, turns on only with the remote flag…
15 entries, 1 of them switched off
Five entries land the reconnection rework for live artifact watches without switching it on. Lease mode, the flag tengu_slate_lantern_ember, defaults to off, so a stock build…
876 entries, 107 of them switched off
A weekly one-shot reset for the five-hour session limit is built but dark: it appears as a hidden /limit-reset command and a menu row only when the server-supplied config…
298 entries, 32 of them switched off
Several features are compiled into this build but cannot be reached yet, and 58 entries fall into that group. A new system-prompt block named willow_tern would tell the model to…
429 entries, 46 of them switched off
This build carries 68 gated or half-landed items, and a dozen are worth naming. Cloud sessions can seed from a plain folder by bundling it in-process, but the step is skipped…
3 entries, 2 of them switched off
Both of this release's visible changes sit around /autofix-pr, the command that watches your current pull request and fixes problems it finds, and neither is switched on. The…
3 entries, 1 of them switched off
This build carries the scaffolding for language-server support, but nothing in it turns that support on. There is a manager with start, restart and shutdown entry points, and none…
396 entries, 65 of them switched off
Several of the 99 unreleased entries center on plugin hook modules written in JavaScript: a plugin's hooks/hooks.json can name a module that registers hooks in code and talks to…
5 entries, 1 of them switched off
Claude Code now sends the VS Code extension a map of experiment flags at startup, and that map picks up an entry called tengu_sable_thrush, which tells the IDE whether narration…
10 entries, 6 of them switched off
This build carries seven pieces of unreleased work, most of it about how the model's thinking is shown. The machinery to pick a thinking display mode is present, but the one path…
300 entries, 44 of them switched off
This build carries 64 features that are present but not reachable, most waiting on server-side flags. If you hit the 5-hour usage limit, a hidden /low-priority command could…
254 entries, 31 of them switched off
A large amount of artifact work sits in this build without a way to switch it on. The verify action, which lets Claude read console output, uncaught errors and failed resource…
Nothing on this page matches that. Press Enter to search every build.