{"version":"2.1.290","summary":"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 a conversation must be before idle compaction starts. A new --proactivity flag picks how much Claude does on its own when it starts. CLAUDE_CODE_DISABLE_PROACTIVITY turns that choice off. Running claude remote-control with --allow-unattended-tool-calls lets cloud sessions use the current folder for one run. Remote sessions now read the SDK address from CLAUDE_CODE_REMOTE_SDK_URL when --sdk-url is not given.\n\nThis release has 7 entries in the build that are not switched on yet. One would let you attach more PDFs as a reference when their page count is known. Another would let an MCP server that needs you to sign in keep its saved tools. Claude Code could also tell the model that you pay per token and what a Bash call costs. Live-update connections could learn to spot a proxy answering in place of the real server. Each of these waits on a switch that is off by default.\n\nAmong this release's fixes, a cancel sent from outside the terminal no longer throws away every prompt you had queued. Plugin names that version 2.1.287 changed now match the installed plugins again. Synced skills now keep the text that sits above their main body. A plugin's hook error handler now still runs if the worker running those hooks is replaced mid-event. Stopping a remotely served background command now returns without waiting for it to end.","filter":{"section":null,"tier":null,"area":null},"offset":0,"limit":50,"count":50,"total":548,"next_offset":50,"next_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/entries.json?offset=50","entries":[{"anchor":"remote-control-gains-allow-unattended-tool-calls-and-a-for","heading":"claude remote-control gains --allow-unattended-tool-calls","glance":"A new remote-control option lets auto-mode cloud sessions run commands in your folder without asking, sandboxed, with status messages showing what is allowed","teaser":null,"tier":"use","area":"Remote Control","section":"What probably matters to you","position":0,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-gains-allow-unattended-tool-calls-and-a-for","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-gains-allow-unattended-tool-calls-and-a-for.json"},{"anchor":"fleet-view-kill-agents-shortcut-hint-and-set-proactivity-le","heading":"A shortcut to stop background agents, and a proactivity control in the SDK","glance":"A hint shows the shortcut that stops background agents, and the SDK gains a request to set the proactivity level","teaser":null,"tier":"use","area":"Background Agents","section":"What probably matters to you","position":1,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/fleet-view-kill-agents-shortcut-hint-and-set-proactivity-le","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/fleet-view-kill-agents-shortcut-hint-and-set-proactivity-le.json"},{"anchor":"new-proactivity-cli-option-hidden-from-help","heading":"A hidden --proactivity option and a matching session control request","glance":"A new `--proactivity <level>` option, hidden from help, sets how much Claude does on its own, and a remote session started from Claude Code is sent the same level","teaser":null,"tier":"use","area":"Proactivity","section":"What probably matters to you","position":2,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-proactivity-cli-option-hidden-from-help","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-proactivity-cli-option-hidden-from-help.json"},{"anchor":"idle-compaction-gets-a-user-setting-and-a-min-token-env-over","heading":"Idle compaction gets an idleCompaction off switch and a minimum-token override","glance":"New idleCompaction setting can stop idle compaction, and CLAUDE_CODE_IDLE_COMPACT_MIN_TOKENS overrides its minimum token threshold","teaser":null,"tier":"use","area":"Compaction","section":"What probably matters to you","position":3,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/idle-compaction-gets-a-user-setting-and-a-min-token-env-over","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/idle-compaction-gets-a-user-setting-and-a-min-token-env-over.json"},{"anchor":"new-proactivity-startup-level-gated-by-tengu-proactivity","heading":"New --proactivity flag sets how much Claude does on its own at startup","glance":"A `--proactivity` flag picks the proactivity level when Claude Code starts, and `CLAUDE_CODE_DISABLE_PROACTIVITY` turns the selector off","teaser":null,"tier":"use","area":"Proactivity","section":"What probably matters to you","position":4,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-proactivity-startup-level-gated-by-tengu-proactivity","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-proactivity-startup-level-gated-by-tengu-proactivity.json"},{"anchor":"claude-code-remote-sdk-url-fallback-for-the-sdk-url-host","heading":"Remote sessions can read the SDK address from CLAUDE_CODE_REMOTE_SDK_URL","glance":"When --sdk-url is not given in a remote session, Claude Code now reads the address from the CLAUDE_CODE_REMOTE_SDK_URL environment variable","teaser":null,"tier":"use","area":"Sessions","section":"What probably matters to you","position":6,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-code-remote-sdk-url-fallback-for-the-sdk-url-host","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-code-remote-sdk-url-fallback-for-the-sdk-url-host.json"},{"anchor":"prompt-string-ablation-via-env-var-or-client-data","heading":"Strings can be removed from prompts with an environment variable","glance":"Setting CLAUDE_CODE_REMOVE_PROMPT_STRINGS to a JSON list removes those strings from the prompts Claude Code sends to the model","teaser":null,"tier":"use","area":"Agents","section":"What probably matters to you","position":7,"scope":"both","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/prompt-string-ablation-via-env-var-or-client-data","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/prompt-string-ablation-via-env-var-or-client-data.json"},{"anchor":"pewter-owl-tool-responses-gated-by-tengu-witty-hearth","heading":"A remote switch now controls an opt-in tool behaviour","glance":"With `CLAUDE_CODE_PEWTER_OWL_TOOL` set, three tools can now leave out a response ID depending on a remotely controlled switch","teaser":null,"tier":"use","area":"Tools","section":"What probably matters to you","position":8,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pewter-owl-tool-responses-gated-by-tengu-witty-hearth","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pewter-owl-tool-responses-gated-by-tengu-witty-hearth.json"},{"anchor":"new-env-var-claude-code-resume-tolerates-context-seeds-chang","heading":"New environment variable changes how resume treats a session&#039;s last message","glance":"Setting CLAUDE_CODE_RESUME_TOLERATES_CONTEXT_SEEDS stops resume treating some final messages as an interrupted turn","teaser":null,"tier":"use","area":"Resume","section":"What probably matters to you","position":9,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-env-var-claude-code-resume-tolerates-context-seeds-chang","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-env-var-claude-code-resume-tolerates-context-seeds-chang.json"},{"anchor":"new-file-descriptor-env-var-for-ccr-agent-proxy-token","heading":"New environment variable for passing the remote agent proxy token","glance":"Claude Code now recognises CCR_AGENT_PROXY_TOKEN_FILE_DESCRIPTOR as a way to receive the agent proxy credential","teaser":null,"tier":"use","area":"Cloud Sessions","section":"What probably matters to you","position":10,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-file-descriptor-env-var-for-ccr-agent-proxy-token","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-file-descriptor-env-var-for-ccr-agent-proxy-token.json"},{"anchor":"new-env-var-claude-code-streamed-bumblebee-text-overrides-st","heading":"New environment variable sets custom wording for the subagent budget note","glance":"Setting `CLAUDE_CODE_STREAMED_BUMBLEBEE_TEXT` replaces the wording of the budget note given to subagents, with `{N}` standing for the budget","teaser":null,"tier":"use","area":"Subagents","section":"What probably matters to you","position":11,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-env-var-claude-code-streamed-bumblebee-text-overrides-st","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-env-var-claude-code-streamed-bumblebee-text-overrides-st.json"},{"anchor":"new-relaunchproactivity-env-vars-added-to-a-scrub-list-no","heading":"New variables to turn off proactivity and carry it across a relaunch","glance":"Claude Code adds `CLAUDE_CODE_DISABLE_PROACTIVITY` and variables that pass proactivity state to a relaunched Claude Code","teaser":null,"tier":"use","area":"Proactivity","section":"What probably matters to you","position":12,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-relaunchproactivity-env-vars-added-to-a-scrub-list-no","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-relaunchproactivity-env-vars-added-to-a-scrub-list-no.json"},{"anchor":"control-request-set-mcp-permission-mode-override-gains-askfo","heading":"Programs hosting Claude Code can force named MCP tools to ask permission","glance":"A host can now list tools that must always ask before running, even ones that would normally allow themselves","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":13,"scope":"both","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/control-request-set-mcp-permission-mode-override-gains-askfo","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/control-request-set-mcp-permission-mode-override-gains-askfo.json"},{"anchor":"bridge-remote-control-peers-can-run-effort-model-rena","heading":"Remote Control users can run \/effort, \/model, \/rename and \/usage while Claude is working","glance":"Someone connected through Remote Control can now run `\/effort`, `\/model`, `\/rename` and `\/usage` straight away while Claude is mid-reply","teaser":null,"tier":"use","area":"Remote Control","section":"What probably matters to you","position":14,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bridge-remote-control-peers-can-run-effort-model-rena","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bridge-remote-control-peers-can-run-effort-model-rena.json"},{"anchor":"nested-subagent-tool-calls-forwarded-to-sdk-consumers-gated","heading":"SDK apps can receive tool calls made by nested subagents","glance":"Programs built on the SDK can now receive tool calls and results from subagents nested inside other subagents","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":15,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/nested-subagent-tool-calls-forwarded-to-sdk-consumers-gated","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/nested-subagent-tool-calls-forwarded-to-sdk-consumers-gated.json"},{"anchor":"new-sdkstream-json-initialize-capabilities-forwardnestedsu","heading":"SDK clients can ask for nested subagent tool calls and a Chrome setup card","glance":"Programs connecting to Claude Code can opt in to tool calls from nested subagents and to showing a Claude in Chrome setup card","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":16,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-sdkstream-json-initialize-capabilities-forwardnestedsu","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-sdkstream-json-initialize-capabilities-forwardnestedsu.json"},{"anchor":"new-system-message-subtype-file-attachments-missing","heading":"Messages with file attachments: missing-file notice and no queue merging","glance":"A new file_attachments_missing message reports attachments that did not arrive, and queued messages with attachments are no longer merged","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":17,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-system-message-subtype-file-attachments-missing","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-system-message-subtype-file-attachments-missing.json"},{"anchor":"per-subagent-total-tokens-reminder-via-tengu-rippling-tulip","heading":"Subagents can get their own token countdown reminder","glance":"A new `CLAUDE_CODE_RIPPLING_TULIP` value sets the token count for the total-tokens reminder, which can be per subagent","teaser":null,"tier":"use","area":"Subagents","section":"What probably matters to you","position":18,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/per-subagent-total-tokens-reminder-via-tengu-rippling-tulip","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/per-subagent-total-tokens-reminder-via-tengu-rippling-tulip.json"},{"anchor":"claude-attach-accepts-a-session-name","heading":"claude attach and claude logs accept a session name","glance":"`claude attach` and `claude logs` now take a background session's name, or part of it, in place of its id","teaser":null,"tier":"use","area":"Sessions","section":"What probably matters to you","position":19,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-attach-accepts-a-session-name","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-attach-accepts-a-session-name.json"},{"anchor":"proactivity-level-injected-as-a-conversation-attachment","heading":"Claude can be told how much to do on its own: low, medium or high proactivity","glance":"A proactivity note in the conversation tells Claude whether to ask first, act normally or act on its own for commits, pushes, pull requests and follow-ups","teaser":null,"tier":"notice","area":"Proactivity","section":"What probably matters to you","position":20,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-level-injected-as-a-conversation-attachment","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-level-injected-as-a-conversation-attachment.json"},{"anchor":"offerchromesetup-tool-added","heading":"New OfferChromeSetup tool lets Claude offer to set up Claude in Chrome","glance":"A new OfferChromeSetup tool lets Claude offer one-time Claude in Chrome setup mid-task when the browser extension is not connected","teaser":null,"tier":"notice","area":"Chrome & Browser","section":"What probably matters to you","position":21,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/offerchromesetup-tool-added","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/offerchromesetup-tool-added.json"},{"anchor":"home-settings-seeded-to-cloud-sessions-ccr","heading":"Your home Claude settings can now be sent to cloud sessions","glance":"With consent, Claude Code now uploads your ~\/.claude CLAUDE.md, rules, output styles and settings to a cloud session when you create or attach to it","teaser":null,"tier":"notice","area":"Cloud Sessions","section":"What probably matters to you","position":23,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/home-settings-seeded-to-cloud-sessions-ccr","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/home-settings-seeded-to-cloud-sessions-ccr.json"},{"anchor":"attention-budget-us-election-info-tool-and-allowunattendedt","heading":"Election info tool name, unattended tool call option and shorter shell error","glance":"Claude Code adds `us_election_info` to its web tool names, an `allowUnattendedToolCalls` option, and trims a shell error message","teaser":null,"tier":"internal","area":"Web Search","section":"What probably matters to you","position":24,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/attention-budget-us-election-info-tool-and-allowunattendedt","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/attention-budget-us-election-info-tool-and-allowunattendedt.json"},{"anchor":"run-and-similar-skills-append-extra-files-and-a-note-from-a","heading":"<code>\/run<\/code> can pick up extra files and a note beyond its bundled ones","glance":"The \/run skill and one other skill now add extra files from another source to their own, and can add a note to the prompt","teaser":null,"tier":"use","area":"Skills","section":"What probably matters to you","position":25,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/run-and-similar-skills-append-extra-files-and-a-note-from-a","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/run-and-similar-skills-append-extra-files-and-a-note-from-a.json"},{"anchor":"compaction-precompact-hook-trigger-reports-auto-vs-manual","heading":"<code>PreCompact<\/code> hooks are now told when compaction was automatic","glance":"The PreCompact hook now receives a trigger of auto for automatic compaction instead of always manual","teaser":null,"tier":"use","area":"Hooks","section":"What probably matters to you","position":26,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/compaction-precompact-hook-trigger-reports-auto-vs-manual","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/compaction-precompact-hook-trigger-reports-auto-vs-manual.json"},{"anchor":"claude-plugin-test-rollout-flag-hold-and-file-form","heading":"<code>claude plugin test<\/code> now waits for its rollout setting to be saved","glance":"`claude plugin test` now waits until a rollout setting is saved for test files to read, and reports when that fails","teaser":null,"tier":"use","area":"Plugins","section":"What probably matters to you","position":28,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-plugin-test-rollout-flag-hold-and-file-form","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/claude-plugin-test-rollout-flag-hold-and-file-form.json"},{"anchor":"remote-control-stop-can-name-which-turn-it-is-for","heading":"A Stop from Remote Control can say which turn it is meant for","glance":"A Remote Control Stop request can now name the messages it targets, and is ignored if none of them belong to the current turn","teaser":null,"tier":"use","area":"Remote Control","section":"What probably matters to you","position":29,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-stop-can-name-which-turn-it-is-for","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-stop-can-name-which-turn-it-is-for.json"},{"anchor":"proactivity-level-can-override-the-starting-permission-mode","heading":"A proactivity level can set the permission mode a session starts in","glance":"When a proactivity level other than default is set, Claude Code can choose the starting permission mode from it","teaser":null,"tier":"use","area":"Permissions","section":"What probably matters to you","position":30,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-level-can-override-the-starting-permission-mode","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-level-can-override-the-starting-permission-mode.json"},{"anchor":"remote-control-turn-end-keeps-subagent-permission-requests-o","heading":"Approval requests from subagents can stay open after a remote turn ends","glance":"In remote sessions, approval requests from subagents can now stay open when a turn ends instead of always being dismissed","teaser":null,"tier":"use","area":"Remote Control","section":"What probably matters to you","position":31,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-turn-end-keeps-subagent-permission-requests-o","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-control-turn-end-keeps-subagent-permission-requests-o.json"},{"anchor":"hook-skipping-notice-for-cloud-session-commands","heading":"Cloud sessions now say when project hooks are skipped on this computer","glance":"Claude Code now tells you when project hooks you haven't accepted are skipped for cloud session commands, pointing to a command to fix it","teaser":null,"tier":"use","area":"Cloud Sessions","section":"What probably matters to you","position":32,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/hook-skipping-notice-for-cloud-session-commands","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/hook-skipping-notice-for-cloud-session-commands.json"},{"anchor":"pr-prep-suggestion-now-requires-includecodereviewsuggestion","heading":"Code review suggestion at commit time now needs a setting","glance":"The code review part of the suggestion shown before a commit now appears only when includeCodeReviewSuggestion is true","teaser":null,"tier":"use","area":"Code Review","section":"What probably matters to you","position":33,"scope":"individual","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pr-prep-suggestion-now-requires-includecodereviewsuggestion","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pr-prep-suggestion-now-requires-includecodereviewsuggestion.json"},{"anchor":"background-subagent-follow-up-foreground-subagents-moved-to","heading":"Foreground subagents move to the background when you queue a message","glance":"With CLAUDE_AUTO_BACKGROUND_TASKS set, running foreground subagents move to the background when you have a queued message waiting","teaser":null,"tier":"use","area":"Subagents","section":"What probably matters to you","position":34,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/background-subagent-follow-up-foreground-subagents-moved-to","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/background-subagent-follow-up-foreground-subagents-moved-to.json"},{"anchor":"managed-settings-snapshotfirst-path","heading":"Managed settings can now be read from a saved snapshot first","glance":"In some cases Claude Code now uses a saved snapshot of managed settings before looking at administrator settings","teaser":null,"tier":"use","area":"Managed Settings","section":"What probably matters to you","position":35,"scope":"org","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/managed-settings-snapshotfirst-path","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/managed-settings-snapshotfirst-path.json"},{"anchor":"set-model-can-now-accept-an-unconfirmed-model-and-returns-u","heading":"Model switches can go through without server confirmation","glance":"A `set_model` request now applies the model even if the server could not confirm it, and the reply says `unconfirmed: true`","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":36,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/set-model-can-now-accept-an-unconfirmed-model-and-returns-u","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/set-model-can-now-accept-an-unconfirmed-model-and-returns-u.json"},{"anchor":"proxyauthhelper-person-and-policy-sources-run-in-the-user-d","heading":"Proxy auth helper runs under new rules for personal and policy sources","glance":"With sources set to personAndPolicy, the proxy auth helper is skipped under folder trust or else runs from another folder","teaser":null,"tier":"use","area":"Network Proxy","section":"What probably matters to you","position":37,"scope":"both","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proxyauthhelper-person-and-policy-sources-run-in-the-user-d","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proxyauthhelper-person-and-policy-sources-run-in-the-user-d.json"},{"anchor":"rewind-conversation-refuse-if-usage-limited-and-allow-summa","heading":"rewind_conversation can reach summarized messages and can refuse when usage-limited","glance":"rewind_conversation gains allow_summarized_target to rewind into compacted history and refuse_if_usage_limited to decline when a usage limit is hit","teaser":null,"tier":"use","area":"SDK","section":"What probably matters to you","position":38,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/rewind-conversation-refuse-if-usage-limited-and-allow-summa","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/rewind-conversation-refuse-if-usage-limited-and-allow-summa.json"},{"anchor":"chrome-menu-reports-live-connection-state-and-reconnect-act","heading":"Better handling when the Claude in Chrome connection fails","glance":"\/chrome now shows the real Claude in Chrome connection state with a working Reconnect, and Claude is told to skip browser tools while it is down","teaser":null,"tier":"use","area":"Chrome & Browser","section":"What probably matters to you","position":39,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/chrome-menu-reports-live-connection-state-and-reconnect-act","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/chrome-menu-reports-live-connection-state-and-reconnect-act.json"},{"anchor":"host-auth-env-var-is-exempted-from-the-environment-scrub","heading":"Variable named by <code>CLAUDE_CODE_HOST_AUTH_ENV_VAR<\/code> is kept when the environment is cleaned","glance":"When Claude Code removes certain environment variables, it now keeps the one named by `CLAUDE_CODE_HOST_AUTH_ENV_VAR`","teaser":null,"tier":"use","area":"Auth","section":"What probably matters to you","position":40,"scope":"both","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/host-auth-env-var-is-exempted-from-the-environment-scrub","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/host-auth-env-var-is-exempted-from-the-environment-scrub.json"},{"anchor":"bundled-claude-api-skill-adds-managed-agents-quickstarts","heading":"Built-in claude-api skill gains managed-agents guides and quickstarts","glance":"The claude-api skill that ships with Claude Code now includes managed-agents reference docs, nine quickstart templates and an onboarding flow","teaser":null,"tier":"notice","area":"Skills","section":"What probably matters to you","position":42,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bundled-claude-api-skill-adds-managed-agents-quickstarts","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bundled-claude-api-skill-adds-managed-agents-quickstarts.json"},{"anchor":"sub-agent-token-budget-note-streamed-bumblebee","heading":"Claude can be told the token budget for each agent it launches","glance":"A note can tell Claude each agent it launches has a fixed token budget, switched on by `CLAUDE_CODE_STREAMED_BUMBLEBEE`","teaser":null,"tier":"notice","area":"Subagents","section":"What probably matters to you","position":43,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/sub-agent-token-budget-note-streamed-bumblebee","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/sub-agent-token-budget-note-streamed-bumblebee.json"},{"anchor":"compaction-summary-prompt-gets-ab-arms-controlleanshort","heading":"Conversation summaries can now be written with one of four prompt styles","glance":"The instructions used to summarize a long conversation now come in four versions, chosen by an environment variable or a server setting","teaser":null,"tier":"notice","area":"Compaction","section":"What probably matters to you","position":44,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/compaction-summary-prompt-gets-ab-arms-controlleanshort","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/compaction-summary-prompt-gets-ab-arms-controlleanshort.json"},{"anchor":"new-elapsed-time-reminder-and-proactivity-attachments","heading":"New elapsed-time and proactivity reminders sent to the model","glance":"Claude Code can now add an elapsed-time reminder and a proactivity note to what the model sees","teaser":null,"tier":"notice","area":"System Reminders","section":"What probably matters to you","position":45,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-elapsed-time-reminder-and-proactivity-attachments","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-elapsed-time-reminder-and-proactivity-attachments.json"},{"anchor":"proactivity-opt-in-restored-across-worker-epochs","heading":"Proactive mode choice now survives worker restarts","glance":"In headless sessions, Claude Code now keeps your proactivity opt-in when the worker restarts","teaser":null,"tier":"notice","area":"Proactivity","section":"What probably matters to you","position":46,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-opt-in-restored-across-worker-epochs","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/proactivity-opt-in-restored-across-worker-epochs.json"},{"anchor":"remote-tools-device-bridge-machine-listings-no-longer-adopt","heading":"Remote machines offered by the device listing can now be set aside","glance":"Claude Code can now decline machines offered by its device listing and take remote machines only from the session's own connection","teaser":null,"tier":"notice","area":"Remote Machines","section":"What probably matters to you","position":47,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-tools-device-bridge-machine-listings-no-longer-adopt","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/remote-tools-device-bridge-machine-listings-no-longer-adopt.json"},{"anchor":"surface-client-device-attestation-floor-enforcement-on-the-r","heading":"Remote sessions can now turn away apps on devices that fall below a trust policy","glance":"Remote sessions now check attached apps against a device trust policy and refuse ones that fall below it","teaser":null,"tier":"notice","area":"Sessions","section":"What probably matters to you","position":48,"scope":"both","heads_up":true,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/surface-client-device-attestation-floor-enforcement-on-the-r","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/surface-client-device-attestation-floor-enforcement-on-the-r.json"},{"anchor":"propose-skills-gains-a-kept-lines-check-gated-by-tengu-glitt","heading":"Skill proposals can be refused when they drop too much of your existing skill","glance":"The skill-proposal tool can now refuse a proposed SKILL.md that keeps too few lines of your current one, using a threshold set remotely","teaser":null,"tier":"notice","area":"Skills","section":"What probably matters to you","position":49,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/propose-skills-gains-a-kept-lines-check-gated-by-tengu-glitt","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/propose-skills-gains-a-kept-lines-check-gated-by-tengu-glitt.json"},{"anchor":"new-gate-tengu-moonlit-kettle-and-a-new-status-icon-set","heading":"Status icons for working, needs you, idle and failed sessions","glance":"Claude Code gains a set of status icons for working, needs-you, idle and failed states","teaser":null,"tier":"notice","area":"Terminal UI","section":"What probably matters to you","position":50,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-gate-tengu-moonlit-kettle-and-a-new-status-icon-set","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/new-gate-tengu-moonlit-kettle-and-a-new-status-icon-set.json"},{"anchor":"subagent-effort-can-be-lowered-by-remote-value-tengu-harmon","heading":"Subagents get their own effort level and token budget","glance":"Subagents now get a computed effort level and total-token budget with a countdown reminder, and a remote value can lower their effort","teaser":null,"tier":"notice","area":"Subagents","section":"What probably matters to you","position":52,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/subagent-effort-can-be-lowered-by-remote-value-tengu-harmon","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/subagent-effort-can-be-lowered-by-remote-value-tengu-harmon.json"},{"anchor":"pdf-reference-attachment-readablewhole-telemetry-under-teng","heading":"PDF reference attachments can tell Claude the whole file is readable","glance":"A new PDF path can attach a PDF as a reference and tell Claude it may read the whole file without picking pages","teaser":null,"tier":"notice","area":"File Attachments","section":"What probably matters to you","position":53,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pdf-reference-attachment-readablewhole-telemetry-under-teng","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/pdf-reference-attachment-readablewhole-telemetry-under-teng.json"},{"anchor":"bash-cost-and-per-token-cost-prompt-sections-added-both-beh","heading":"Prompt notes on token cost and Bash call cost, both off by default","glance":"Claude Code can tell the model the user pays per token and what a Bash call costs, but both notes fall back to off","teaser":null,"tier":"soon","area":"System Reminders","section":"What probably matters to you","position":54,"scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bash-cost-and-per-token-cost-prompt-sections-added-both-beh","entry_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.290\/e\/bash-cost-and-per-token-cost-prompt-sections-added-both-beh.json"}],"facets":{"sections":[{"value":"What probably matters to you","entries":61},{"value":"Improvements","entries":335},{"value":"Bug Fixes","entries":37},{"value":"In Development","entries":5},{"value":"Internal Changes","entries":107},{"value":"Removed","entries":3}],"tiers":[{"value":"notice","entries":387},{"value":"internal","entries":118},{"value":"use","entries":36},{"value":"soon","entries":7}],"areas":[{"value":"Cloud Sessions","entries":47},{"value":"Permissions","entries":46},{"value":"MCP","entries":38},{"value":"Sessions","entries":36},{"value":"Plugins","entries":30},{"value":"Remote Control","entries":28},{"value":"Elsewhere","entries":21},{"value":"Telemetry","entries":21},{"value":"Terminal UI","entries":20},{"value":"Hooks","entries":17},{"value":"Background Tasks","entries":14},{"value":"Git","entries":14},{"value":"Internals","entries":13},{"value":"SDK","entries":13},{"value":"Skills","entries":13},{"value":"Subagents","entries":12},{"value":"Chrome & Browser","entries":10},{"value":"Worktrees","entries":10},{"value":"Auth","entries":9},{"value":"Messaging","entries":8},{"value":"File Attachments","entries":7},{"value":"Resume","entries":7},{"value":"Sandbox","entries":7},{"value":"System Reminders","entries":7},{"value":"API Errors","entries":6},{"value":"Compaction","entries":6},{"value":"Proactivity","entries":6},{"value":"Agent Teams","entries":5},{"value":"Channels","entries":5},{"value":"File Paths","entries":5},{"value":"Settings","entries":5},{"value":"Startup","entries":5},{"value":"Auto Mode","entries":4},{"value":"Model Switching","entries":4},{"value":"Plan Mode","entries":4},{"value":"Scheduled Tasks","entries":4},{"value":"Shell","entries":4},{"value":"Web Search","entries":4},{"value":"Artifacts","entries":3},{"value":"GitHub","entries":3},{"value":"Secret Redaction","entries":3},{"value":"Streaming","entries":3},{"value":"Tools","entries":3},{"value":"CLAUDE.md","entries":2},{"value":"Code Review","entries":2},{"value":"Context Management","entries":2},{"value":"Gateway","entries":2},{"value":"Web Fetch","entries":2},{"value":"Agents","entries":1},{"value":"Background Agents","entries":1},{"value":"Effort","entries":1},{"value":"File Editing","entries":1},{"value":"Localization","entries":1},{"value":"Managed Settings","entries":1},{"value":"Network Proxy","entries":1},{"value":"Remote Machines","entries":1}]}}