Skip to main content

AI Tools Changelog Hub

The latest updates from 36 AI tools at a glance

36 Tools•2,246 recorded updates•Last sync: Oct 22, 05:04 AM•RSS

Search and timeline show 147 of 2,246 recorded updates. The hub loads only a recent selection per tool.

Today

Piv1.4.0New

v1.4.0

October 22, 2026

  • •Image input for classifiers: codemode models.classify() accepts images for TypeSafe's Jev and Cloudflare's Clef models on typesafe, cloudflare-workers-ai, and OpenRouter. See Use classifier models and Codemode.
  • •Models Configuration reference: every provider, model, model override, and compatibility field of models.json, generated from its schema. See Models Configuration.
  • •Summaries without a fixed output cap: compaction and branch summaries use the model's output limit, so long reasoning no longer makes them fail. See Compaction.
  • •Codemode waits for a free worker: parallel scripts and in-process subagents queue instead of failing, and PI_CODEMODE_MAX_WORKERS raises the limit. See Codemode limits.
  • •Rex terminal support: OSC 8 hyperlinks, truecolor, and inline images.
  • •Added a Models Configuration reference generated from the models.json schema, listing all provider, model, model override, and compatibility fields (#10790)
  • •Codemode models.classify() now accepts images for TypeSafe's Jev and Cloudflare's Clef models on typesafe, cloudflare-workers-ai, and OpenRouter, and lists Clef Omni and Jev 1.14 on Workers AI.
  • •Provider-level compat in models.json now also applies to classifier models, with the System One fields imageInput and sendImageOptInHeader for proxies that serve another service's classifiers. Extension classifier models accept compat too.
  • •Added MiniMax-M3.1-Flash to the minimax and minimax-cn providers, with thinking levels low through max (#11121)
  • •Added support for the Rex terminal: OSC 8 hyperlinks, truecolor, and inline images (#11131, #11150 by @petrroll, #11183 by @nihaals)
  • •Provider attribution headers now come from pi-ai's client identity. Extensions' before_provider_headers handlers no longer see the OpenRouter, NVIDIA NIM, and OpenCode attribution headers; returning null for one still removes it. Cloudflare requests send pi's detailed User-Agent instead of pi-coding-agent, as do GitHub release checks for tool downloads, and OpenCode requests send x-opencode-client whenever attribution is enabled, not only with a session.
  • •generateSummary() and generateSummaryWithUsage() now ignore their reserveTokens parameter. Summary requests use the model's output limit instead.
  • •Changed the default GitHub Copilot model from gpt-5.4, which Copilot retired, to gpt-6.1-sol.
  • •Changed the default NVIDIA model to nvidia/nemotron-3.5-super-vl-120b-a12b (#10910 by @cv)
  • •Changed the default Xiaomi model to mimo-v2.6-pro for the xiaomi and Xiaomi Token Plan providers, since mimo-v2.5-pro left the catalog.
Claude Agent SDKv0.3.307New

v0.3.307

October 22, 2026

  • •Added wakeup_source to the command_lifecycle started frame of a scheduled wakeup or task, and scheduled_for for a wakeup (capability lifecycle_wakeup_source_v1), so a host can tell which one fired
  • •Added supportedDialogKinds to browser SDK query(), sent on initialize; a non-empty list requires onUserDialog
  • •Updated to parity with Claude Code v2.1.307

This Week

OMPv18.12.6New

v18.12.6

October 21, 2026

  • •Added the prompt-cache-persistent catalog axis (isPromptCachePersistent) for providers whose own endpoint never expires prompt caches; logged-in OpenAI Codex declares it
  • •Updated the bundled Yolo-Auto catalog to its current models, qwen3.8-27b (now the default), yolo, yolo-small, and yolo-128k, and removed the retired deepseek-flash-v4 and qwen3.8-flash
  • •Added the _session/steering ACP extension (advertised as _meta.steering.supported), so clients such as agent-shell can steer a running turn; a steer sent while a permission dialog is open cancels that dialog, and _meta.steering.queue: "followUp" queues a follow-up instead (#15885 by @marcintustin)
  • •/mcp reauth <name> now works over ACP, RPC, and print mode (omp -p '/mcp reauth <name>'): it opens the browser, prints the authorization URL, reconnects the server in the running session, and print mode exits 1 when reauthorization fails (#16046 by @marcintustin)
  • •ACP session/cancel now also cancels prompts that are queued or still dispatching, and stops autonomous runs, instead of being ignored when no client turn is streaming (#15885 by @marcintustin)
  • •Idle compaction no longer runs on logged-in OpenAI Codex accounts, whose prompt cache never expires; compacting while idle only threw away a warm cache
  • •ACP sessions no longer fail to start when a client-declared MCP server cannot connect; the failure is reported in the session, and an HTTP/SSE server can be repaired with /mcp reauth (#16046 by @marcintustin)
  • •/mcp reauth, /mcp test, and /mcp unauth now pick the project entry when a server is defined in both project and user config, and skip project config when mcp.enableProjectConfig is off (#16046 by @marcintustin)
  • •/mcp reauth removes the server's stale Authorization header from its config when it writes the new OAuth binding (#16046 by @marcintustin)
  • •omp -p '/compact', /handoff, and /retry given as CLI arguments now run the command instead of sending the text to the model, and exit 1 when the command fails; slash text arriving on stdin or from @file is still sent to the model (#16387 by @marcintustin)
  • •Interactive startup no longer waits for extensions that register only tools, commands, flags, renderers, or event hooks (for example MCP bridge plugins): they load once the composer accepts input, and your first message waits for them instead. An extension is deferred only after one launch has confirmed it registers no providers, shortcuts, composer shapes, or file fallbacks. Launches that resume a session, pass --tools, or pass flags omp does not recognize still load every extension first (#16684 by @H4vC)
  • •Fixed subagents whose role has an explicitly empty fallback chain falling back through the default chain when the default role uses the same model, including after a retry fallback, prewalk or context promotion briefly switched models or the subagent was revived (#16548 by @eggpeat)
  • •Fixed /usage and omp usage --json listing identity-less OAuth accounts such as Devin as missing usage beside their working reports (#16549 by @eggpeat)
  • •Fixed releasing or reaping a browser tab returning while Chromium still listed the page; the close now waits until Chromium reports the tab destroyed (#16680 by @H4vC)
  • •Fixed the advisor prompt forbidding all repeated advice, including real escalations: advisors may now re-raise a prior nit or concern as a blocker once new evidence makes it one (#13887, #16484 by @pjh890321)
Claude Codev2.1.307New

Claude Code 2.1.307

October 21, 2026

  • •Added /claude-api api-version to move Claude API code to API version 2026-10-21
  • •Added the claude_code.safety_event OpenTelemetry event for each tool call a permission rule, auto mode, a safety check, a hook or the sandbox stopped, prompted or sent back
  • •Fixed requests failing with a model-not-found error when CLAUDE_CODE_DISABLE_1M_CONTEXT is set and the model is an alias with the [1m] suffix, such as opus[1m]
  • •Fixed the prompt-caching-scope beta header being sent behind custom ANTHROPIC_BASE_URL gateways and on Foundry, where it has no effect
  • •Fixed MCP servers showing "needs authentication" in claude -p on macOS while the login keychain was locked; they now say why, and the next run after an unlock connects without a new sign-in
  • •Fixed MCP servers being reported as timed out, and staying failed in headless runs, when a macOS or Linux computer went to sleep while they were connecting
  • •Fixed HTTP MCP tool calls being sent twice when the connection closed mid-call; only read-only or idempotent tools are now retried
  • •Fixed a turn sometimes starting without an SDK MCP server's latest tools when a background task was running
  • •Fixed some "Autocompact is thrashing" errors in turns that read many PDF pages, images or other large tool results
  • •Fixed /autocompact in cloud sessions going back to the default window after the session restarted
  • •Fixed a conversation cleared with /clear sometimes coming back when a cloud or self-hosted runner session restarted
  • •Fixed CLAUDE_STREAM_FIRST_BYTE_TIMEOUT_MS being ignored in Claude apps gateway sessions
  • •Fixed telemetry from Claude Desktop sessions behind a Claude apps gateway being rejected when OTEL_EXPORTER_OTLP_HEADERS is set
  • •Fixed /permissions > Recently denied not showing the denied shell command, and auto mode's local classifier ignoring an approval from there when it checks a retried command; approving a row now names the command, not its description
  • •Fixed /help and the / command menu showing long model-facing descriptions for bundled skills such as /claude-api instead of their short descriptions
  • •Fixed subagent transcripts recording placeholder token usage and an empty stop reason for streamed responses
  • •Fixed model switches through the Agent SDK's setModel(), and apps built on it, being refused when the API took more than 5 seconds to confirm the model; Claude Code now waits up to 15 seconds
  • •Fixed a background command stopped through a connected app (such as the desktop app archiving an idle session) being reported to Claude as stopped by the user; the notice now names its output file
  • •Fixed WebFetch in Cowork cloud sessions reporting a tool error when a site answered with a 4xx status or a robots.txt refusal; it is now a normal result, as in local sessions
  • •Fixed write tools on a directly configured GitHub MCP server showing a form that closed without creating anything in the desktop app's Code tab
  • •Fixed Artifact tool reads failing with a connection reset when the server had closed a reused connection
  • •Self-hosted runner: Fixed sessions with --use-anthropic-git-proxy committing a .claude/settings.local.json left in the clone by an earlier session
  • •Self-hosted runner: Fixed archiving or requeueing a session, or stopping the runner, during its checkout hook leaving the hook's processes running and the runner's slot held
  • •Improved recovery from unparseable tool calls: Claude Code now retries the call up to twice before ending the turn with an error
  • •Improved fork subagents launched together: each is now told which launch is its own, so forks are much less likely to try to start each other's work
  • •Improved browser sign-in: the success page now appears as soon as sign-in completes, without waiting for your account profile to load
  • •Improved web search errors in cloud and self-hosted runner sessions: a search refused with a 403 that gives a reason now shows that reason instead of a generic "Authentication error"
  • •Improved the time to the first reply in cloud sessions when one of the desktop app's MCP servers can't connect
  • •Improved Remote Control connect time for sessions whose transcript contains large pasted images or PDF reads
  • •Changed the notice Claude gets when the WebSearch budget is used up to also tell it not to query search engines in a browser or through computer use
  • •Changed the axScreenReader setting and CLAUDE_AX_SCREEN_READER in a settings env block to apply only from user, managed or --settings settings (since 2.1.305)
  • •Self-hosted runner: Changed checkout hooks to be stopped after 30 minutes by default, failing the session's setup; if yours needs longer, raise --checkout-hook-timeout-min or set it to 0
  • •Self-hosted runner: Changed the certificate file built for Anthropic-managed git to filter its trusted certificates as the CLI does; a [runner:warn] line gives the number left out
  • •[VSCode] Added a notice on the microphone button when voice dictation stops by itself (15 seconds without recognized speech, or the 2-minute limit), saying why and that the text so far is kept
  • •[VSCode] Fixed a long prompt expanded with Show more covering the whole chat panel: a pinned prompt is now capped at 40% of the panel and scrolls
  • •[VSCode] Fixed part of a list item or table cell with LaTeX math pasting as a new list or table in rich text editors
  • •[VSCode] Fixed dictation failing behind a TLS-inspecting proxy when NODE_EXTRA_CA_CERTS is set only in ~/.claude/settings.json; it now also follows CLAUDE_CODE_CERT_STORE
  • •[VSCode] Changed the "still open somewhere else" notice to also cover a Claude process left over from an extension restart, which usually ends on its own within about 10 minutes
  • •[VSCode] Changed voice dictation to filter its trusted certificates as the CLI does
  • •[VSCode] Changed the extension's own requests to filter their trusted certificates as the CLI does, and to trust the NODE_EXTRA_CA_CERTS file named in the environment or in user settings
  • •[Cloud sessions] Fixed the repository picker saying "No repositories match" while a search was still running; it now shows "Searching…" until the results arrive
  • •[Cloud sessions] Fixed the "Waiting for a runner" status on self-hosted environments being cut off; with on-demand runners it now counts up from your last message and shows failed attempts
  • •[Code Review] Added @claude review stop (also off and pause) to Code Review, ending an @claude review always subscription on a pull request without starting a new review
Hermes Agentv0.22.4New

Hermes Agent v0.22.4

October 21, 2026

  • •Desktop no longer shows suggestion pills above the composer. The pills appeared for no clear reason, mostly nudging you to set up MCP servers. They and the helpers that only they used are gone. (#142559)
  • •Plugins that install nothing skip the dependency consent step. A plugin that declares no dependencies, or whose pyproject.toml installs nothing, now enables without an empty consent prompt. (#142569)
  • •Dashboard: WhatsApp can be safely re-paired after logging out. (#142439)
  • •Desktop plugin review: the surface lint no longer flags prototype writes; reviewers check them by hand. (#142586)
  • •Desktop: the approval-mode chip follows the active profile's mode. (#142565)
  • •Release builds: the desktop bundle's pre-build check gets enough time to verify archived build inputs when storage is slow. (#142623)
  • •Microsoft Store app: updates through the Store.
  • •Desktop app installed from the installer (.exe / .dmg): no change in this release; it updates with the next bundled release.
  • •CLI: hermes update, or re-run the installer one-liner.
  • •Docker / Hermes Cloud: nousresearch/hermes-agent:stable (or :latest).
Hermes Agentv0.22.3New

Hermes Agent v0.22.3

October 21, 2026

  • •Microsoft Store app: updates through the Store.
  • •Desktop app installed from the installer (.exe / .dmg): no change in this release; it updates with the next bundled release.
  • •CLI: hermes update, or re-run the installer one-liner.
  • •Docker / Hermes Cloud: nousresearch/hermes-agent:stable (or :latest).
OpenClaw2026.10.10-beta.3New

openclaw 2026.10.10-beta.3

October 21, 2026

  • •Windows reliability: restore fresh state initialization and Gateway startup, hide SQLite helper windows during app updates, and make forced Gateway reinstalls tolerate launcher timestamp settling. (#176194, #174969, #176174, #176945; related #176111, #174906, #176163, and #176872) Thanks @skyzhao1223, @RomneyDa, @easyteacher, @throwingbones, @tjarrau-dev, and @cnnceart.
  • •Upgrade recovery: converge app-owned plugins to the upgraded core version, restore affected agent stores including held system agents after schema-changing updates, and scope migration rollback backups to the versioned tables being changed. (#174257, #175007, #176783, #172671; related #174120 and #176669) Thanks @RomneyDa, @kazuyuki-eguchi, and @PennyVibe.
  • •Agent state: concurrent initialization can no longer let two agents claim ownership of the same agent database. (#174251)
  • •Codex and native apps: previews no longer block replies after a context refresh, session replies resume after native child recovery, bound Codex history survives app-server restarts, and native app messages send again after reconnecting. (#177716, #177564, #172705, #171528) Thanks @jjjhenriksen and @RomneyDa.
  • •Background tasks: completions are no longer dropped when the config changes mid-turn. (#169529; related #169054) Thanks @wangmiao0668000666, @RomneyDa, and @Oldrich333.
  • •Discord: restore native slash commands in channels and threads. Thanks @montrellcruse and @RomneyDa.
  • •Agents: keep the prompt cache when a rate-limited run retries, and preserve steering messages that arrive without an explicit target. Thanks @montrellcruse, @RomneyDa, and @obviyus.
  • •Cron/CLI continuity: preserve CLI chat sessions across cron announce wakes instead of resetting their provider session. (#176218; related #176208) Thanks @eliavb43 and @RomneyDa.
  • •CLI tools: invalidate reused CLI sessions when the available tool catalog changes, restoring tools such as exec after Gateway restarts and preventing stale catalogs from reaching spawned agents. (#176577; related #175670) Thanks @RomneyDa, @jackylecalvez2, and @vincentkoc.
  • •GMI: omit unsupported reasoning_effort values on GPT-5.6 tool turns so the provider no longer rejects them with HTTP 400. (#176709) Thanks @RomneyDa.
  • •Windows updates: recognize Scheduled Task permission denials in every Windows display language, keep Windows plugin dependency links across Doctor phases, and clean up backup scratch directories without EPERM failures. (#178624, #178219, #169672, #176836; related #176743) Thanks @RomneyDa, @n572770, @xutongxue233, and @WG-Mojo.
  • •Update and Doctor: update finalization honors TMPDIR for disposable copies, Doctor suggests a retry instead of a restore after snapshot contention, and repeated Docker container restarts no longer take a full Doctor pre-repair capture each time. (#178009, #176779, #173711; related #176750 and #173564) Thanks @RomneyDa, @WG-Mojo, and @8bitsforge.
  • •Memory: cold Gateway starts no longer open every memory index (restoring 2026.10.4 startup memory), memory files keep indexing after plugin reloads, and plugin reloads no longer time out during slow dreaming cleanup. (#174718, #167756, #169945) Thanks @RomneyDa and @Stronggenetics.
  • •Nodes: synchronous node exec output is no longer repeated as a separate Exec finished wake after the waiting caller already received it. (#164511) Thanks @holgergruenhagen and @obviyus.
  • •macOS: long exec commands through the macOS app node no longer fail after 20 seconds. (#172117; related #139706) Thanks @holgergruenhagen, @RomneyDa, @ylcn91, and @Colton-Harris.
Claude Agent SDKv0.3.306New

v0.3.306

October 21, 2026

  • •Updated to parity with Claude Code v2.1.306
OMPv18.12.5New

v18.12.5

October 21, 2026

  • •Added OpenAI 12-hour prompt caching for GPT-5.6+ on the Responses API: cacheRetention: "long" now requests prompt_cache_options.ttl: "12h" (instead of the ignored legacy prompt_cache_retention: "24h"), and promptCache.ttl accepts "12h"
  • •Fixed every OpenAI Codex request failing with HTTP 404 {"detail":"Not Found"} on personal ChatGPT accounts whose token is not region-pinned (#16493)
  • •Fixed Moonshot kimi-k3 pricing cache writes at $0; usage cost now includes cache writes at Moonshot's $3/M 5-minute rate (#16508)
  • •providers.cacheRetention: long now keeps OpenAI GPT-5.6+ prompt caches for 12 hours (Responses API; cache writes cost 2.5× input instead of 1.25×)
  • •Fixed subagents spawning at a caller-supplied effort (e.g. hi → max) while task.enableEffort is off; the field is now ignored and the agent's configured thinking level (including :auto) applies (#16402)
  • •Fixed omp update --check failing against JFrog Artifactory registries that add empty dist-tags to a tagged manifest; the update check now uses the registry's curated latest version (#16375)
  • •Fixed <path>:conflicts on a file past the 10 MiB scan cap reporting no conflicts with no hint: it now names the line the scan stopped at and the ranged read that registers the blocks after it (#16269)
  • •Fixed the ts-set-map rule stalling tool-call streaming for seconds on a long single-line Map literal, and missing Map entries written ["key", value] as const (#16242)
  • •Fixed the ts-import-type rule missing inline type imports whose member name is non-ASCII, such as import("./units").温度 (#16242)
  • •Fixed a macOS background action leaving its target in front of your app after you clicked another of your windows, when placing the click took longer than the click itself (#16217)
  • •fix(update): handled empty Artifactory dist-tags on tagged manifests by @roboomp in https://github.com/can1357/oh-my-pi/pull/16379
  • •fix(task): ignored caller effort while task.enableEffort is off by @roboomp in https://github.com/can1357/oh-my-pi/pull/16406
  • •fix(catalog): stopped declaring no_constraint codex residency by @roboomp in https://github.com/can1357/oh-my-pi/pull/16501
  • •fix(catalog): priced moonshot kimi-k3 cache writes at the 5min rate by @roboomp in https://github.com/can1357/oh-my-pi/pull/16529
Claude Codev2.1.306New

Claude Code 2.1.306

October 21, 2026

  • •Fixed claude -p inside cloud sessions reporting "Not logged in" when started without the session's environment variables
  • •Fixed an API error being replaced by "prompt is too long", and an unneeded compaction, in sessions with many MCP tools
Droid (Factory AI)v0.246.0New

CLI v0.246.0, Desktop v0.203.0: Chat image gallery navigation, faster transcripts, authentication progress, browser panel safety, and CLI and app fixes

October 21, 2026

  • •New features: Chat image gallery - Step through the images of one message in the full-screen preview with arrow keys, buttons, or dots
  • •Improvements: Session transcript rendering - Long session transcripts now stay responsive while the session page updates
  • •Improvements: Credits in usage notices - The high-cost model notice and limits modal now describe usage in credits instead of tokens
  • •Improvements: Planning status for actions - Droid now shows one planning status with the number of actions instead of a row per guessed call
  • •Improvements: Fetching public URLs - Droid can now read any public URL with FetchUrl instead of only URLs you supplied
  • •Bug fixes: Buy Extra Usage notice - The Buy Extra Usage modal now asks for a payment method instead of promising to charge a card
  • •Bug fixes: Download save dialog - Downloading a file from a remote session now opens the save dialog right away
  • •Bug fixes: Questions written as text - Droid now asks its question when a model writes the request as plain text instead of showing markup
  • •Bug fixes: Unarchive button size - The Unarchive button in an archived session's bar now matches the size of the other controls
  • •Bug fixes: Selector menu icons - Icons in the project, branch, and machine selector menus now sit centered in their rows
  • •Bug fixes: Authentication progress - A session now shows when it is authenticating instead of looking idle
  • •Bug fixes: Terminal tab title - The terminal tab now keeps the current session's title instead of taking one from another session
  • •Bug fixes: Slack session titles - A session started from Slack no longer shows raw mention codes in its title
  • •Bug fixes: Desktop sign-in retries - The desktop app now retries a connection instead of reporting a failed sign-in when verification is unreachable
  • •Bug fixes: Subagents on Auto model - A subagent left on Auto now keeps the tools its chosen model needs instead of running without them
Droid (Factory AI)v0.245.1New

CLI v0.245.1, Desktop v0.202.1: Personal GitLab Self-Hosted accounts for organization members

October 21, 2026

  • •New features: Personal GitLab Self-Hosted accounts - Members can connect their own account on an organization's self-hosted GitLab, and managers can require it for repository access
Gemini CLIv0.65.0New

Release v0.65.0

October 20, 2026

  • •fix(cli): enforce read-only workspace settings in untrusted folders by @jvargassanchez-dot in https://github.com/google-gemini/gemini-cli/pull/29583
  • •fix(core): avoid duplicate tool response turns when resuming sessions by @diegogodinezr in https://github.com/google-gemini/gemini-cli/pull/29618
  • •fix(core): prevent deletion of resumed session history on quick exit (#29198) by @villahernandez-coder in https://github.com/google-gemini/gemini-cli/pull/29584
  • •Changelog for v0.64.0-preview.0 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/29656
  • •Changelog for v0.63.0 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/29659
  • •fix(core): align OAuth callback iss parameter validation with RFC 9207 metadata by @luisfelipe-alt in https://github.com/google-gemini/gemini-cli/pull/29616
  • •fix(cli): prevent unnecessary terminal clears and scroll resets when expanding with Ctrl+O by @jesussamuel-byte in https://github.com/google-gemini/gemini-cli/pull/29640
  • •chore(release): bump version to 0.65.0-nightly.20261006.gfb972b2f8 by @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/29657
  • •fix(ci): add missing loop in unassign-inactive-assignees workflow by @ugorla-dev in https://github.com/google-gemini/gemini-cli/pull/29609
  • •fix(core): enforce terminal user turn invariant and normalize request contents by @luisfelipe-alt in https://github.com/google-gemini/gemini-cli/pull/29612
  • •feat(telemetry): support custom OTLP headers in telemetry configuration by @jesussamuel-byte in https://github.com/google-gemini/gemini-cli/pull/29641
  • •fix(auth): prevent infinite verification and OAuth retry loops by @villahernandez-coder in https://github.com/google-gemini/gemini-cli/pull/29655
  • •fix(cli): handle JSON parse and response stream errors in fetchJson by @jesussamuel-byte in https://github.com/google-gemini/gemini-cli/pull/29658
  • •fix(core): preserve line terminators in truncateString by @diegogodinezr in https://github.com/google-gemini/gemini-cli/pull/29673
  • •fix(vscode-ide-companion): make IdeServer.stop() resolve while MCP sessions are open by @elberthc-byte in https://github.com/google-gemini/gemini-cli/pull/29674
Claude Agent SDKv0.3.305New

v0.3.305

October 20, 2026

  • •Added recommendations and fix to the get_usage response's behavior windows
  • •Fixed servers set with setMcpServers being replaced later in the session by a same-named server from the user's own settings
  • •Fixed startup() with initializeTimeoutMs: null counting time spent in SessionStart or Setup hooks against its 60-second limit; null now works like leaving the option out
  • •Changed the startup() and prewarm() timeout error for a start with no open SessionStart or Setup hook: it lists possible causes and no longer says to check authentication and network
  • •Updated to parity with Claude Code v2.1.305
OMPv18.12.4NewBreaking Change

v18.12.4

October 20, 2026

  • •Enabled native compaction for supported Bedrock Claude models (#16343 by @anatoli-tsinovoy)
  • •Added opt-in reclaimAbovePct to auth.accountPolicies: sessions automatically moved to a lower-priority account return to the preferred one once it has that much quota left (#14237 by @will-bogusz)
  • •Added reachesBudgetedHost(): whether the catalog's host request budgets (inline image bytes, official image count) apply to a request, true on a model's official endpoint and on Factory Droid's own hosts (#16288 by @will-bogusz)
  • •Added Inception API-key login (/login inception) and INCEPTION_API_KEY environment authentication (#16439 by @eggpeat)
  • •Added image support for native OpenAI Decisions and Workers AI judgments, prompted vision chat, and auth-gateway System One requests; images exceeding a judgment API's documented caps (e.g. Clef's 4 images, 4 MiB each) are rejected locally (#15471 by @anatoli-tsinovoy).
  • •isSurvivingAssistant in providers/transform-messages: whether an assistant turn reaches a replay, malformed tool calls dropped first, so callers outside transformMessages pick the same latest turn (#15791 by @will-bogusz)
  • •Broker account-pool files (OMP_AUTH_BROKER_ACCOUNT_POOL_FILE) can hide whole providers and API keys: "*": false makes the file an allow-list, and an entry can be true, false, or { "accounts": [...], "apiKeys": false } (#14235 by @will-bogusz)
  • •cursorRawReadPath() writes a bare start line as raw:N- (file:10 → file:raw:10-), so Cursor exec reads of such a path still read from line N onward now that read takes N:raw as one line (#16438 by @will-bogusz)
  • •Fixed Anthropic and Bedrock logging one warning per dropped thinking block (hundreds for one request) after a conversation change; the drop now logs one warning with the count (#16392 by @will-bogusz)
  • •Fixed Factory Droid Core models (Kimi, GLM, DeepSeek and others) dropping calls to tools with names over 64 characters and their results from later requests; such names now shorten to 64 characters, and shorter names are sent unchanged (#16384 by @will-bogusz)
  • •Fixed Devin /usage showing no data for accounts signed in with /login devin, and Devin chat sending a failing extra auth request for legacy Windsurf Enterprise keys (#13677 by @will-bogusz)
  • •Turns stopped for a thinking loop now keep the tokens and cost the provider had reported, the response id, the request's start time, and its time to first token and duration, instead of recording zero usage and no timing (#16437 by @will-bogusz)
  • •Fixed Claude requests answering a turn whose attachment is a video, PDF, HEIC or other format sent outside a JPEG/PNG/GIF/WebP image as if nothing were attached; the turn now fails with an error naming the format (#16407 by @will-bogusz)
  • •Fixed the repeated-tool-call guard missing a loop the model interleaves with a wait that elapses with nothing new; a wait that returns finished work still ends the repetition (#11874 by @dickyudhandika)
  • •Fixed Google Antigravity chat and image requests failing with false 429 quota-exhausted errors (#11849 by @jotafurtado)
Droid (Factory AI)v0.245.0New

CLI v0.245.0, Desktop v0.202.0: Chrome cookie import, folder links in the Files panel, voice dictation answers, and CLI and app fixes

October 20, 2026

  • •New features: Chrome cookie import - The built-in browser can import your Chrome cookies on macOS, so you stay signed in to your sites
  • •New features: Folder links in chat - Clicking a folder path in Droid's message now opens the Files panel and reveals that folder
  • •Improvements: Voice dictation answers - Droid's questions now take a dictated answer, and the dictation level meter has a new design
  • •Improvements: Remote session footer - The footer now separates a remote session's location from its branch with a dot
  • •Improvements: Thinking label shimmer - The shimmer on the thinking label now lines up with the text it covers
  • •Bug fixes: MCP tools with union inputs - An MCP tool whose input schema uses a top-level union now works instead of failing
  • •Bug fixes: Blocked request errors - A refused request now shows a readable message instead of an empty error
  • •Bug fixes: Saved project in New Session - New Session now shows your saved project right away instead of waiting for your computers to load
  • •Bug fixes: Resuming remote sessions - Resuming a remote session no longer delays your messages while it loads that machine's scheduled runs
  • •Bug fixes: Reserved MCP server names - Adding an MCP server under a reserved name now reports an error instead of saving nothing
  • •Bug fixes: Wiki repository URL matching - The wiki commands now find a repository's wiki when its URL is written in another form
  • •Bug fixes: Plugin MCP server paths - A plugin's MCP server now resolves its plugin folder instead of reporting an unset variable
  • •Bug fixes: Newly added GitHub repositories - Repositories added to your GitHub connection now show up in the repository picker
  • •Bug fixes: Linux in-app updates - Updating the desktop app on Linux now unpacks the release instead of failing
  • •Bug fixes: Stuck background commands - Droid no longer keeps waiting on a command whose process has already exited
Claude Agent SDKv0.3.304New

v0.3.304

October 19, 2026

  • •Added suspendSseStream(query) and resumeSseStream(query) to the browser SDK: close a query's event stream and reopen it later without ending the query
  • •Added stable row ids to getContextUsage(), and, when the CLI answers on its own stdio, plugin ids, agent and skill file paths, and per-server MCP instruction token shares
  • •Fixed startup() timing out while a SessionStart or Setup command hook was running when no initializeTimeoutMs is passed: up to 10 minutes of that time no longer counts toward the 60-second limit, and the timeout error now says what the process reported
  • •Fixed startup() keeping every hook_progress message of a running SessionStart or Setup hook in memory until it resolved; it now keeps only the newest one for each hook
  • •Updated to parity with Claude Code v2.1.304
OpenClaw2026.10.10-beta.2New

openclaw 2026.10.10-beta.2

October 19, 2026

  • •Workspace safety: reject a multi-file apply_patch before changing files when a later update target is missing or its context does not match, preserving the original workspace for the next attempt. Commit-time I/O failures can still leave partial effects. (#160409; related #117742 and #127496) Thanks @masatohoshino, @obviyus, and @yetval.
  • •Tool policy: run plugin before_tool_call hooks in standalone openclaw agent exec sessions so plugin tool guards take effect. (#159333; fixes #158824) Thanks @joseairosa and @obviyus.
  • •Buzz: reconnect when a relay closes during NIP-42 authentication without terminating the Gateway or interrupting other channels and sessions. (#170591; fixes #147269) Thanks @masatohoshino, @obviyus, @jarvis-modern-mail, and @RomneyDa.
  • •macOS downloads: save dashboard and conversation attachments through the native Save dialog while keeping the current page open. (#173654) Thanks @Patrick-Erichsen.
  • •OpenRouter onboarding: accept valid API keys with the default openrouter/auto model instead of rejecting setup with a route mismatch. (#173292; related #162798) Thanks @chandraveshchaudhari and @obviyus.
  • •macOS setup: let fresh native profiles start the Gateway and complete automatic setup, including when supervised startup needs a retry. (#173246)
  • •Gateway startup: follow a replacement model-runtime refresh instead of exiting when initial preparation is superseded. (#173366; related #173357) Thanks @Jackten and @RomneyDa.
  • •Gateway logging: keep serving and shut down normally when a deleted temporary-directory backing path makes daily log rollover fail. (#160433; fixes #130807) Thanks @masatohoshino, @obviyus, @WadeOwenWatts, and @Patrick-Erichsen.
  • •Gateway stability: stop session diff capture from opening live SQLite and hard-linked files, preventing SIGBUS crash loops when the shared-memory file is truncated. (#174017)
  • •Native app updates: let Doctor finish agent database upgrades on native-first profiles instead of reporting success while the Gateway refuses to start on an older schema. (#174297)
  • •Cron: plan history maintenance from run identities and load full payloads only for rows it updates, so large cron histories no longer exhaust the SQLite worker's memory and take channels down. (#174530; fixes #174078) Thanks @wangmiao0668000666, @RomneyDa, and @asikorskiy.
  • •Windows Claude CLI: stop every turn with a stored Anthropic credential from stalling until the no-output watchdog when Claude Code never reads the credential pipe. (#174571; fixes #174204) Thanks @RomneyDa and @scalesmart.
  • •Plugins: restore the SDK facade exports that published external plugins still import, so those plugins load again. (#173777; fixes #173386) Thanks @RomneyDa and @FtlC-ian.
  • •Chat tools: keep the chat usable after removing a tool server that failed to start, cleaning up its failed startup before reporting the launch error. (#171397) Thanks @joelagnel.
  • •Updates: bound the post-activation Doctor step by the update's allowance instead of waiting indefinitely with the Gateway down. (#173862)
Claude Agent SDKv0.3.303New

v0.3.303

October 19, 2026

  • •Updated to parity with Claude Code v2.1.303
OpenClaw2026.10.5New

openclaw 2026.10.5

October 18, 2026

  • •Release notes — formatted for people.
  • •Changelog — plain Markdown for AI agents and tools.
  • •npm package: https://www.npmjs.com/package/openclaw/v/2026.10.5
  • •registry tarball: https://registry.npmjs.org/openclaw/-/openclaw-2026.10.5.tgz
  • •integrity: sha512-9rCuXMweqVKZIK3KBtGcLzesz9G6+FmQ30lvwomRhmcNWDcOtlBoqdLbPditoJMuTP55jPqXpt1AA0L8tSzFHg==
  • •release SHA: 1b58cb9a1b4f4353e1add63c8506db4d890962e8
  • •full release CI report: https://github.com/openclaw/releases/blob/main/evidence/2026.10.5/release-evidence.md
  • •release publish: https://github.com/openclaw/openclaw/actions/runs/39167610006
  • •npm preflight: https://github.com/openclaw/openclaw/actions/runs/39149250969
  • •full release validation: https://github.com/openclaw/openclaw/actions/runs/39148728797
  • •plugin npm publish: https://github.com/openclaw/openclaw/actions/runs/39168400620
  • •plugin ClawHub submission: https://github.com/openclaw/openclaw/actions/runs/39168403212; public finalization and exact artifact verification follow terminal release-parent completion
  • •plugin ClawHub bootstrap: not needed
  • •OpenClaw npm publish: https://github.com/openclaw/openclaw/actions/runs/39164633114/attempts/1
  • •npm Telegram beta E2E: not supplied
Orcav1.4.229New

Inline chat review, shared skills for more agents, and connected Jira planning

October 17, 2026

  • •Review and share Native Chat work — inspect an agent's edits turn by turn in the conversation, expand diagrams into a pan-and-zoom view, share chat visuals with a click, and keep Claude goals and Codex context usage in view
  • •Shared skills for more agents — install shared skills across 20 additional agent tools, with broader WSL discovery and the right install locations for Kiro, Qoder, CodeBuddy, and more
  • •Jira planning on desktop and phone — switch between issue lists and boards, focus on the current sprint, reuse saved JQL and Jira favorites, start workspaces from either device, and let agents read and update Jira tickets through the Orca CLI
Claude Agent SDKv0.3.302New

v0.3.302

October 16, 2026

  • •Fixed structured output results delivering a response's output as structured_output when the API refused that response; the result is now an error with no payload
  • •Changed an omitted permissionMode to start in default mode on third-party providers, with telemetry off or on a fresh install; pass permissionMode: 'auto' to keep auto mode
  • •Changed fork subagents (subagent_type: "fork") to be on by default in SDK sessions, as in the terminal; run_in_background is still offered, and CLAUDE_CODE_FORK_SUBAGENT=0 turns them off
  • •Updated to parity with Claude Code v2.1.302
OpenCodev1.19.3New

v1.19.3

October 16, 2026

  • •Configure model reasoning options to generate reasoning variants, replacing catalog variants while preserving your configured variant overrides.
Clinev4.1.25New

Cline 4.1.25

October 16, 2026

  • •Cline Cloud sessions, rolling out gradually: run a task in a hosted sandbox on a GitHub repo and branch you pick. Turn it on in Settings. Cloud tasks show in Recent and History with their status, notify you when they finish, and keep running after you close your editor.
  • •With no model chosen, the Cline provider now uses Gemini 4 Argon, the new first model in Cline's recommended list. OpenRouter stays on Claude Sonnet 5.5.
  • •Refreshed the model catalog. Adds the Boundless and Standard Compute providers, and SambaNova now comes from the catalog with MiniMax M3 as its default. The Cline recommended list adds Gemini 4 Argon and drops Grok 4.7, and the Cline free list replaces Step 5 Preview with Nemotron 3.5 Super VL. OpenCode Go no longer lists MiniMax M2.7. Default models change for Cortecs, CrossModel, Eden AI, Google Gemini, GitHub Copilot, Google Vertex AI, LLM Gateway, Ofox, OpenCode Go, and Prime Inference (Gemini 4 Argon), Fireworks AI, Nebius Token Factory, and Requesty (Muse Spark Open), CoreWeave, DigitalOcean, NanoGPT, and Nvidia (Nemotron 3.5 Super VL), Aixy and Neon (Claude Haiku 5.5), Agnes AI, Ambient, IteraCompute, Poe, Scaleway, and SenseNova. If you use one of those providers without pinning a model, expect a different default.
  • •Long requests to local models (LM Studio, Ollama) no longer fail after 5 minutes of silence with terminated: BodyTimeoutError while the server is still processing a large prompt.
  • •OpenAI Compatible base URLs are used exactly as entered. A host-only URL like https://abc.com used to get /v1 added silently. Host-only URLs saved by earlier versions get /v1 added once, so existing setups keep working.
  • •Images and files attached to an answer to one of Cline's questions now reach the model instead of being dropped.
  • •When a provider error interrupts a response mid-stream, Cline retries it up to 3 times before failing the turn.
  • •In JetBrains, the Run task card stays above the Auto-approve bar.
  • •Auto Compact and its strategy setting sit together again in Feature settings.
OpenCodev1.19.2New

v1.19.2

October 16, 2026

  • •OpenCode 1 now stays on the 1.x release line: update checks use only the npm opencode-ai latest version, so 1.x installs are never offered a 2.x build as a regular update.
  • •Upgrading to OpenCode 2 remains a separate, explicit choice through the 2.0 offer in /update or opencode upgrade --major.
  • •Fix upgrades for installs managed by yarn.
  • •Skip 1.x updates for Homebrew Core installs, which now track OpenCode 2, and check tap and Chocolatey availability before offering an update.
  • •The install script refuses to install 2.x versions.
OpenClaw2026.10.4New

openclaw 2026.10.4

October 16, 2026

  • •The Gateway keeps running after the computer sleeps or the system clock jumps, instead of restarting itself because its ownership heartbeat looked stale. (#170790) Thanks @AnvinaLI, @sgh6688, @waynegault, and @RomneyDa.
  • •Gateway shutdown and restart no longer hang on a stalled transcript worker or in-flight session lifecycle events. (#170493, #170555) Thanks @Oldrich333.
  • •Tlon group DMs no longer treat other participants as the owner, closing an owner-impersonation path. (#165801) Thanks @eleqtrizit.
  • •Headless browser pages keep responding to clicks while a site waits for a passkey. (#170857) Thanks @trukhinyuri.
  • •Manual /compact settles the active run before compacting instead of aborting it and dropping the pending request. (#166163) Thanks @holny and @alexph-dev.
  • •Stability patch for 2026.10.3 with no new features or configuration changes. Thanks @RomneyDa.
  • •Gateway: stays responsive during long reply-tagged streams, and long replies that quote tool-call XML no longer re-parse Markdown on every token. (#166782, #171573) Thanks @boeingchoco.
  • •Gateway: blocked durable context-engine turns are logged once instead of on every run. (#167595) Thanks @jalehman.
  • •Control UI: delayed application heartbeats no longer force a reconnect. (#171253)
  • •Gateway: recover state ownership after sleep and wall-clock changes, and allow openclaw gateway restart from verified external macOS terminals. (#170790, #170354) Thanks @AnvinaLI, @sgh6688, @waynegault, and @RomneyDa.
  • •Gateway: bound transcript-worker shutdown waits and wait for session lifecycle publication during shutdown. (#170493, #170555) Thanks @Oldrich333.
  • •Gateway: start enabled standalone plugin files with the Gateway, and keep worker heap limits enforced when the managed service sets a process-wide heap flag. (#171330, #166581) Thanks @holgergruenhagen, @valkyriweb, @xydigitLybnnnn, @goslingmanagment, @waynegault, and @obviyus.
  • •Agents: launched plugin turns no longer fail with writer-claim errors, and Claude CLI sessions no longer fork history across message routes. (#170077, #167241) Thanks @cmatrx0.
  • •Agents: manual compaction settles the active run instead of aborting it. (#166163) Thanks @holny and @alexph-dev.
  • •OpenAI: tool turns for reasoning models on api.openai.com Completions routes no longer fail with a 400. (#166182) Thanks @obviyus and @AsTutoAI.
Cursor2026-10-16New

Cloud Agent Canvases and Environment Improvements

October 16, 2026

  • •Learn more about sharing canvases.
  • •The Environments API is now available in beta. Use it to programmatically create and manage the cloud environments your agents run in.
  • •Cloud agent environments now support up to 300 repositories. Large environments also load 5x faster.
Droid (Factory AI)v0.244.0New

CLI v0.244.0, Desktop v0.201.0: Offline session archiving, session arrangement tiles, signed macOS download, and CLI and app fixes

October 16, 2026

  • •New features: Session arrangement tiles - New sessions now open in an empty arrangement slot, and you can drag dividers to resize tiles
  • •Improvements: Welcome screen title - The welcome screen now leads with its title instead of a decorative label above it
  • •Improvements: Update download progress - Running an update now shows how much of the download has finished
  • •Improvements: Late connector tools - A turn that began before your connector or MCP tools finished loading now continues once they are ready
  • •Bug fixes: Archiving offline sessions - Archiving a session on a disconnected machine now succeeds right away, and that machine catches up when it reconnects
  • •Bug fixes: Signed macOS download - The macOS desktop app download is now signed and served from a stable link
  • •Bug fixes: Organization-required review workflows - A required review workflow whose rule names a branch now runs Droid instead of being refused
  • •Bug fixes: Markdown table columns - Narrow columns in a markdown table now keep their words whole when the table shrinks
  • •Bug fixes: Plan status replies - Droid no longer shows a plan status line in the transcript when no work is left
  • •Bug fixes: Replies after interrupted tools - A message sent right after you interrupt a tool now gets an answer instead of another tool call
  • •Bug fixes: Spec comment box - Starting a comment on a long spec now keeps the spec and the comment box in view
Droid (Factory AI)v0.243.0New

CLI v0.243.0, Desktop v0.200.0: Browser pane input, upload cancellation, large file reads, sign-in, and required review fixes

October 16, 2026

  • •Bug fixes: Machine details after reload - A reloaded desktop window now picks up the local machine's details instead of leaving them unknown (app)
  • •Bug fixes: Browser pane keyboard input - Droid's keystrokes now reach the browser pane they target, and fail clearly when it cannot take focus (app)
  • •Bug fixes: Multi-select custom answer - The custom answer row in a multi-select question now shows a checkbox that checks itself once you type (app)
  • •Bug fixes: Canceling file uploads - An attachment that is still uploading can now be canceled from the chat box (app)
  • •Bug fixes: Network failure toasts - Returning to the app after sleep no longer repeats failed request toasts for background refreshes (app)
  • •Bug fixes: Pattern kill commands - A kill by pattern no longer ends the command that ran it
  • •Bug fixes: Sign-in across tabs - Opening the web app no longer signs you out when another browser tab is stuck refreshing your session
  • •Bug fixes: Reading very large files - Droid now reads a range of a very large file without running out of memory
  • •Bug fixes: Background rm commands - A background command that deletes a file with rm -f is no longer refused
  • •Bug fixes: Pull request in footer - The footer now updates its pull request when the session changes folder
  • •Bug fixes: Organization-required Droid reviews - A review workflow required across an organization now runs Droid on new pull requests
  • •Bug fixes: Subagent transcript scrolling - The subagent transcript popup now keeps following new output as it arrives (app)
Codex CLIv0.163.0New

0.163.0

October 15, 2026

  • •Open /config to search preferences across categories, change supported settings immediately, and see which changes require a restart. (#52990, #53429)
  • •Send a follow-up while Codex is responding to steer it immediately. Interrupting a turn also stops active Code Mode cells and their nested tool calls. (#51812, #51835)
  • •Configure a leader key in /keymap. The default is Ctrl+X, shortcut hints stay open until you choose or cancel, and voice mute moves to F9. (#52273)
  • •Restart a managed background server from /daemon. Codex warns about work the restart could interrupt, then reconnects automatically. (#53173)
  • •Codex can run independent read-only tools in parallel, including skill, memory, and conversation-history lookups. (#52245)
  • •Daybreak is enabled by default for eligible accounts. codex exec can also select it with an OpenAI API key when the server offers access. (#53701, #53717)
  • •Resuming a saved thread restores its permissions and network approvals, while respecting explicit overrides and current managed policy. (#53150, #53419, #53424)
  • •The network proxy now checks permission before a DNS lookup, rejects duplicate Host headers, and prevents credential aliases from bypassing inspection hooks. (#51650, #52661, #53506)
  • •Return to an existing side conversation with /side after switching threads. Answers to pending prompts also go back to the thread that asked. (#52184, #53436)
  • •On Windows, Ctrl+V and Shift+Insert paste text as well as images. On WSL, repeated pastes reuse the clipboard reader, and copy attempts time out instead of hanging. (#53224, #52967, #53453)
  • •Local commands finish promptly when a child keeps an output pipe open. Remote commands preserve late-arriving output so the final poll can still collect it. (#52652, #53331, #53622)
  • •codex exec no longer prints the same file diff again after an unrelated command. (#53444)
  • •Updated the bundled OpenAI Docs guidance for migrations to the GPT-6 family, including model-specific reasoning, tool use, and data residency considerations. (#51768)
  • •Clarified how the Linux sandbox uses bubblewrap for filesystem isolation and seccomp for syscall filtering. (#52610)
  • •Codex now uses its standard shell executor and no longer bundles a patched zsh. Configurations using the retired shell flags receive a migration notice. (#52160, #52213)
Claude Agent SDKv0.3.301New

v0.3.301

October 15, 2026

  • •Added per-tool permission policies to sdk-type MCP server configs: tools entries with permission_policy always_ask or always_deny, applied on initialize and setMcpServers()
  • •Added BUILTIN_TOOL_NAMES and LEGACY_TOOL_NAME_ALIASES exports: the public built-in tool names and their earlier names, for checking permission rules and tool settings
  • •Added addPlugins to Query.reloadPlugins() so a host can load more plugin folders into a running session
  • •Fixed setModel() with a model ID failing while the account is at its five-hour or weekly usage limit; the model now applies and the next message reports the limit
  • •Changed the error when a full disk keeps the login from being refreshed: error is now invalid_request, not authentication_failed, and api_error_status is null, not 401
  • •Changed mcpServerStatus() to still report a tool's _meta.ui.visibility when the rest of its _meta.ui fails validation
  • •Updated to parity with Claude Code v2.1.301
OpenCodev1.19.1New

v1.19.1

October 15, 2026

  • •Session sharing has been discontinued. Existing share links and session-sharing commands are no longer available, and importing sessions from share URLs is unsupported.
Piv1.3.0New

v1.3.0

October 15, 2026

  • •Choose where `/share` uploads: sessionShareDestination selects auto, gist, or radius. See Sessions and context and Export or share a session.
  • •Cache-preserving thinking level changes on OpenAI GPT-6: changing the thinking level mid-session with an API key keeps the prompt cache.
  • •Kimi Code over Chat Completions: tools added mid-conversation no longer invalidate the prompt cache. models.json base URL overrides must now include /v1. See Providers.
  • •Clean model switching: thinking from a previous model is dropped instead of being shown to the new model as text.
  • •Resumed sessions keep their active tools: --tools +name/-name applies on top of them. See Tools.
  • •Changing the thinking level mid-session on OpenAI GPT-6 models (API key) now keeps the prompt cache (#10853)
  • •Added the sessionShareDestination setting to choose automatic, GitHub gist, or Radius uploads for /share (#10870 by @vegarsti)
  • •Changed the kimi-coding provider to use OpenAI Chat Completions at https://api.kimi.com/coding/v1, the protocol used by the official Kimi Code client. K3, K3-256K, and Kimi For Coding now accept tools added mid-conversation without invalidating the prompt cache. models.json overrides of the kimi-coding base URL must now include /v1 (#10720)
  • •Changed model switching to drop thinking blocks from earlier assistant messages produced by a different model, instead of sending them as plain text. The new model previously saw the old model's reasoning as part of its replies and could start writing its own reasoning into the visible answer (#10995)
  • •Pinned internal Pi dependencies to exact release versions to prevent fresh installs from mixing incompatible packages (#10865)
  • •Fixed resumed sessions dropping tools activated during the session. Resume now restores the session's active tools, with --tools +name/-name entries applied to them, unless --tools with tool names, --no-tools, or --no-builtin-tools selects tools explicitly (#10785)
  • •Fixed stopping a codemode script or reaching its deadline waiting for a non-interruptible built-in to return on Bun (#10748)
  • •Fixed repeated codemode timeouts or aborts accumulating surviving workers; admission now limits live workers to two across sessions, counting cancelled workers until actual exit (#10748)
  • •Fixed --model choosing a provider by catalog order when several providers offer the match; Pi now prefers the provider with credentials and otherwise reports the ambiguity. --models and enabledModels add a model for each provider that offers it and warn about the ambiguity (#10160)
  • •Fixed read line-range labels for numeric-string or null arguments in the TUI, session tree, HTML exports, and examples (#9887)
Hermes Agentv0.22.2New

Hermes Agent v0.22.2

October 15, 2026

  • •The gateway starts after a reboot again. After a restart, Linux can hand an early PID to an unrelated process, and the gateway could mistake it for a running gateway, so a supervised gateway (systemd, launchd, Docker) refused to start. Hermes now records the boot it started in, and when it can't tell, it asks the host lock, which only a live gateway holds. (#138594, fixes #133744)
  • •Long local-model turns survive on laptops. Local requests always stream with prompt-reading progress (shown in the status line), stall timers skip time the machine was asleep, a recovered local server gets another try instead of a breaker that failed every later message, and compression no longer times out while the server is still reading a long conversation. On Windows unified-memory machines, model launch and the prompt cache are sized to free memory. (#138560)
  • •Release: split the Microsoft Store bundle into its own windows-store lane (`f9d8a740`) — @ethernet8023
  • •Agent: add nemotron super 3.5 default parameters (`c59d2994`) — @elisam0
  • •Cli: add nemotron to NVIDIA NIM default model list (`cc55270e`) — @elisam0
  • •Cli: add curated model list at the top in the model picker list (`e6835fd3`) — @elisam0
  • •Release: choose what a stable release builds and tests with lanes (`36a8ee18`) — @ethernet8023
  • •Plugin-catalog: add hermes-lang-ko — Korean Desktop language pack (`e5f49b6a`) — @dandacompany
  • •Plugin-catalog: add hermes-lang-pt-br (Brazilian Portuguese language pack) (`0169e80c`) — @StudioCodeAI
  • •Contributors: add 2867297362mjb-sketch email mapping (`9ab40641`) — @2867297362mjb-sketch
  • •Plugin-catalog: add teamorouter (TeamoRouter model providers) (`ef223406`) — @2867297362mjb-sketch
  • •Plugin-catalog: add author-feed (`5f6d6836`) — @TakeshiTGAL
  • •Plugin-catalog: bump aux-ledger to 1.4.0 (`6048066e`) — @olric3015
  • •Plugin-catalog: add the wait-ledger catalog card (`77cb58e5`) — @olric3015
  • •Plugin-catalog: add wait-ledger (`97e65322`) — @olric3015
OpenAI Agents SDKv0.24.0

v0.24.0

October 15, 2026

  • •expose MCP annotations on converted tools (#5356) (75813f8)
  • •support run-wide tool exception formatting (#5326) (8fd9a7c)
  • •batch OpenAI conversation session writes within the API limit (#5334) (26345c1)
  • •defer automatic compaction for pending approvals (#5363) (5eb47a5)
  • •deps-dev: bump fastapi from 0.139.2 to 0.141.1 (#5312) (cdf23bd)
  • •deps-dev: bump grpcio from 1.83.1 to 1.84.0 (#5310) (a85487e)
  • •deps-dev: bump pymdown-extensions from 11.0.2 to 12.1 (#5314) (d9758b0)
  • •deps-dev: bump pymongo from 4.18.1 to 4.18.2 (#5313) (858856f)
  • •deps: bump boto3 from 1.42.75 to 1.43.109 (#5378) (7da6a92)
  • •deps: bump filelock from 3.20.3 to 4.0.12 (#5375) (14f4553)
  • •deps: bump fsspec from 2025.7.0 to 2026.6.0 (#5317) (66504e9)
  • •deps: bump redis from 7.0.1 to 8.1.0 (#5379) (24ebf1c)
  • •deps: bump temporalio from 1.33.0 to 1.34.0 (#5377) (c148280)
  • •deps: bump vercel-sandbox from 0.7.0 to 0.8.0 (#5374) (2938d87)
  • •deps: update fastapi requirement from &gt;=0.141.1 to &gt;=0.142.2 in /examples/realtime/twilio_sip (#5372) (f47d2c3)
Droid (Factory AI)v0.242.0

CLI v0.242.0, Desktop v0.199.0: Ask Side Chat about highlighted text, resume a session by id, and CLI and app fixes

October 15, 2026

  • •New features: Ask Side Chat about text - Highlighting Droid's response now offers Ask side chat, which opens the side chat with the text quoted
  • •New features: Resume a session by id - The new /resume command, and /sessions, now open a session directly when you pass its id
  • •New features: Continue a spec in chat - An approved spec can now continue in a new chat that starts with the spec's context
  • •Bug fixes: Request timeout retries - A request that times out is now retried instead of immediately failing your turn
  • •Bug fixes: Custom model output limits - Requests to a custom model now stay within the output limit configured for it
  • •Bug fixes: Output after a model switch - A turn that switches models no longer keeps the first model's unfinished reasoning or tool calls
  • •Bug fixes: Browser pane auto-open - The desktop browser pane now opens only after Droid loads a page, and stays closed briefly after you dismiss it
  • •Bug fixes: Secret scan time limit - Scanning a long push history now gets more time before the scan blocks the command
  • •Bug fixes: Connector logos while scrolling - Connector logos in a tool summary no longer paint over the pinned message while you scroll
  • •Bug fixes: Deleted Slack message error - A request started from a Slack message that was deleted now returns a clear error
  • •Bug fixes: Long responses cut short - A response cut off mid-stream is now retried so the turn can finish
  • •Bug fixes: Duplicate usage reports - Usage reported for your own model key is no longer recorded twice when a report is retried
  • •Bug fixes: Right pane key presses - Pressing Enter on a right pane control no longer answers a pending question in chat
  • •Bug fixes: Automation folder icons - A project group in the sidebar now shows the automation folder icon only when every session is an automation
  • •Bug fixes: Process kill commands - Droid now refuses a kill command that would signal its own process
Orcav1.4.226

Jira planning boards, experimental WSL hosts, and interactive Mermaid diagrams

October 15, 2026

  • •Jira planning boards and workspace workflows — Jira integration adds boards, sprint and status views, saved filters, planning columns, prompt templates, and safe CLI access for agents; folder workspaces now appear in the ⌘J jump palette, and new workspaces can start with a setup script alone
  • •Interactive diagrams and richer Native Chat — Mermaid diagrams expand into a pan-and-zoom viewer, Codex chats display context usage, each chat keeps its own permission mode across model switches, and terminal conversations from Grok, OpenCode, and OMP can resume in a new Native Chat
  • •Experimental WSL managed hosts — run a managed Orca host directly inside a WSL distribution with adopted background processes, faster wake paths, and console-free wsl.exe execution
Hermes Agentv0.22.1

Hermes Agent v0.22.1

October 15, 2026

  • •Agent): add nemotron super 3.5 default parameters (`c59d2994`) — @elisam0
  • •Cli): add curated model list at the top in the model picker list (`e6835fd3`) — @elisam0
  • •Plugin-catalog): add hermes-lang-ko — Korean Desktop language pack (`e5f49b6a`) — @dandacompany
  • •Plugin-catalog: add hermes-lang-pt-br (Brazilian Portuguese language pack) (`0169e80c`) — @StudioCodeAI
  • •Plugin-catalog: add teamorouter (TeamoRouter model providers) (`ef223406`) — @2867297362mjb-sketch
  • •Plugin-catalog): add author-feed (`5f6d6836`) — @TakeshiTGAL
  • •Plugin-catalog): bump aux-ledger to 1.4.0 (`6048066e`) — @olric3015
  • •Plugin-catalog): add the wait-ledger catalog card (`77cb58e5`) — @olric3015
  • •Plugin-catalog): add wait-ledger (`97e65322`) — @olric3015
  • •Community plugin: lightningai + lightningai-rate-limit (`07515466`) — @dinosauria123
  • •Catalog: add Hermes AAC 0.1.0 at reviewed exact source pin (`7c9adda8`) — @TREE-Ind
  • •Plugin-catalog): add openinvest (`b6e86101`) — @longsizhuo
  • •Catalog: add hermes-agent-mixer (v0.2.0 @ 04c4e13) (`2e7e3859`) — @Sahil-SS9
  • •Plugins): move Kumiho Memory to memory catalog shelf (`46f95e08`) — @kaveone
  • •Add canonical group attachment controls (`4673f17c`)

This Month

Claude Agent SDKv0.3.300

v0.3.300

October 14, 2026

  • •Fixed importSessionToStore skipping transcript rows that contain a Unicode line or paragraph separator (U+2028, U+2029) on Node 24 and Bun
  • •Fixed a background subagent's final report in auto mode being joined into one turn with another message from the same subagent; the report now always runs as its own turn
  • •Fixed rewind_conversation with interrupt_if_running stopping the running turn before refusing with "unseen later turn"
  • •Reverted canUseTool options carrying suppressAlwaysAllowRule: true for connector tools that an organization set to require approval (added in 0.3.292)
  • •Improved the browser SDK bundle (@anthropic-ai/claude-agent-sdk/browser): 40% smaller to download, by no longer bundling a Node crypto polyfill
  • •Changed thinking: { type: 'disabled' } and maxThinkingTokens: 0 to run as adaptive thinking on models that can't turn thinking off
  • •Changed updatedInput from PreToolUse and PermissionRequest hook callbacks: it is now validated like Claude's own input and unknown keys are dropped; canUseTool is unchanged
  • •Updated to parity with Claude Code v2.1.300
herdrv0.9.5

v0.9.5

October 14, 2026

  • •Prefix shortcuts work again when the terminal sends the next key as plain text. This affected Tern, Chinese, Japanese and Korean input methods on macOS, and Ghostty text: keybinds that send the prefix and a key together. Workspace navigation had the same problem. (#5255)
  • •Codex sessions no longer start with a "clamping Interrupt hook timeout" warning. Herdr no longer installs Codex turn hooks, and Codex status comes from screen detection again. Run herdr integration install codex to update. (#5246, #5278)
  • •Claude Code questions show as blocked in named sessions. (#4573, #5192)
  • •On Windows, plugin panes start even when a folder on PATH has the same name as the plugin's program. (#5267, #5269, thanks @JJLiebig)
  • •On Windows, Claude Code session tracking works when the user folder name contains non-ASCII characters, such as Korean. (#5253, #5270, thanks @JJLiebig)
  • •Adding a Windows machine no longer prints raw PowerShell progress output. (#4639, #5282, thanks @JJLiebig)
Showing 40 of 147

Frequently Asked Questions about the Changelog Hub

Answers to the most important questions about our AI Tools Changelog Hub

Our Changelog Hub currently tracks 36 AI tools and agents: Claude Code, Antigravity, Cursor, Codex CLI, Codex App, OpenCode, OpenClaw, Hermes Agent, Droid (Factory AI), Claude Cowork, Muse, Grok Bot, Grok Build, Muse Code, OMP, Pi, Orca, herdr, Amp, Gemini CLI, Aider, Windsurf (Devin Desktop), VS Code + Copilot, Cline, Roo Code, Continue, Sourcegraph Cody, Zed, Amazon Q Developer, Claude Desktop, ChatGPT Desktop, Claude.ai, ChatGPT, LangChain, Claude Agent SDK, and OpenAI Agents SDK.

When the directory snapshot is refreshed, we fetch new releases for Claude Code, Codex CLI, OpenCode, OpenClaw, Hermes Agent, OMP, Pi, herdr, Gemini CLI, Cline, Roo Code, Claude Agent SDK, and OpenAI Agents SDK from GitHub and for Antigravity, Cursor, Codex App, Droid (Factory AI), Claude Cowork, and Orca from official changelog pages. Muse, Grok Bot, Grok Build, Muse Code, Amp, Aider, Windsurf (Devin Desktop), VS Code + Copilot, Continue, Sourcegraph Cody, Zed, Amazon Q Developer, Claude Desktop, ChatGPT Desktop, Claude.ai, ChatGPT, and LangChain have no reliable dated release feed, so we review and add entries manually from official announcements.

Curated entries are written in both German and English. New external releases may initially appear in English until a German translation is available. You can switch languages in the top right corner.

A Breaking Change is a modification that may impact existing functionality. This can include changed APIs, removed features, or incompatible updates. We clearly mark such changes so you know when you need to adjust your workflow.

Yes! Each changelog page offers an RSS feed. You can find the link in the stats bar next to the tool and update count. This way you can stay up to date without having to visit the page.

We welcome feedback! If you notice an important AI tool is missing or find errors in a changelog, please let us know. We will review and update the data accordingly.

Changelog

The latest updates and improvements to our Changelog Hub
v1.2.0January 11, 2026

FAQ and Changelog Added

  • FAQ section with frequently asked questions about the Changelog Hub
  • Changelog section for hub updates
  • Improved page structure
v1.1.0January 10, 2026

Enhanced Changelog Display

  • Complete changelog history instead of only 6 entries
  • Expand/collapse for long changelog entries
  • Header and footer on all changelog pages
  • Bugfixes for multiple tool changelogs and detail pages
v1.0.0January 9, 2026

Initial Launch

  • Changelog Hub with a curated tool set
  • GitHub Releases API integration
  • Official release and changelog sources for multiple tools
  • Automatic German translation with GPT-5-nano
  • Timeline and grouped view
  • Search and filter by tool, breaking changes, major releases
  • RSS feed for all tools
  • Redis caching for optimal performance