chore: Update CHANGELOG.md and feed.xml

Replace the static ANTHROPIC_API_KEY secret with Workload Identity
Federation inputs in claude.yml, claude-issue-triage.yml, and
claude-dedupe-issues.yml. The federation rule, organization, service
account, and workspace IDs are read from repository variables.

.github/workflows/claude-dedupe-issues.yml

@@ -17,6 +17,8 @@ jobs:
     permissions:
       contents: read
       issues: write
+      # Required to mint the OIDC token exchanged for a Claude API access token (Workload Identity Federation)
+      id-token: write
 
     steps:
       - name: Checkout repository
@@ -31,7 +33,13 @@ jobs:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           allowed_non_write_users: "*"
           prompt: "/dedupe ${{ github.repository }}/issues/${{ github.event.issue.number || inputs.issue_number }}"
-          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          # Authenticate to the Claude API via Workload Identity Federation
+          # (the workflow's OIDC token is exchanged for a short-lived access
+          # token) instead of a static API key.
+          anthropic_federation_rule_id: ${{ vars.ANTHROPIC_FEDERATION_RULE_ID }}
+          anthropic_organization_id: ${{ vars.ANTHROPIC_ORGANIZATION_ID }}
+          anthropic_service_account_id: ${{ vars.ANTHROPIC_SERVICE_ACCOUNT_ID }}
+          anthropic_workspace_id: ${{ vars.ANTHROPIC_WORKSPACE_ID }}
           claude_args: "--model claude-sonnet-4-5-20250929"
 
       - name: Log duplicate comment event to Statsig

.github/workflows/claude-issue-triage.yml

@@ -18,6 +18,8 @@ jobs:
     permissions:
       contents: read
       issues: write
+      # Required to mint the OIDC token exchanged for a Claude API access token (Workload Identity Federation)
+      id-token: write
 
     steps:
       - name: Checkout repository
@@ -34,6 +36,12 @@ jobs:
           github_token: ${{ secrets.GITHUB_TOKEN }}
           allowed_non_write_users: "*"
           prompt: "/triage-issue REPO: ${{ github.repository }} ISSUE_NUMBER: ${{ github.event.issue.number }} EVENT: ${{ github.event_name }}"
-          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          # Authenticate to the Claude API via Workload Identity Federation
+          # (the workflow's OIDC token is exchanged for a short-lived access
+          # token) instead of a static API key.
+          anthropic_federation_rule_id: ${{ vars.ANTHROPIC_FEDERATION_RULE_ID }}
+          anthropic_organization_id: ${{ vars.ANTHROPIC_ORGANIZATION_ID }}
+          anthropic_service_account_id: ${{ vars.ANTHROPIC_SERVICE_ACCOUNT_ID }}
+          anthropic_workspace_id: ${{ vars.ANTHROPIC_WORKSPACE_ID }}
           claude_args: |
             --model claude-opus-4-6

.github/workflows/claude.yml

@@ -33,6 +33,12 @@ jobs:
         id: claude
         uses: anthropics/claude-code-action@v1
         with:
-          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
+          # Authenticate to the Claude API via Workload Identity Federation
+          # (the workflow's OIDC token is exchanged for a short-lived access
+          # token) instead of a static API key.
+          anthropic_federation_rule_id: ${{ vars.ANTHROPIC_FEDERATION_RULE_ID }}
+          anthropic_organization_id: ${{ vars.ANTHROPIC_ORGANIZATION_ID }}
+          anthropic_service_account_id: ${{ vars.ANTHROPIC_SERVICE_ACCOUNT_ID }}
+          anthropic_workspace_id: ${{ vars.ANTHROPIC_WORKSPACE_ID }}
           claude_args: "--model claude-sonnet-4-5-20250929"
 

CHANGELOG.md

@@ -1,10 +1,46 @@
 # Changelog
 
+## 2.1.150
+
+- Internal infrastructure improvements (no user-facing changes)
+
+## 2.1.149
+
+- `/usage` now shows a per-category breakdown of what's driving your limits usage — skills, subagents, plugins, and per-MCP-server cost
+- `/diff` detail view can now be scrolled with the keyboard (arrows, `j`/`k`, `PgUp`/`PgDn`, `Space`, `Home`/`End`)
+- Markdown output now renders GFM task list checkboxes (`- [ ] todo` / `- [x] done`) instead of plain bullets
+- Enterprise: added the `allowAllClaudeAiMcps` managed setting to load claude.ai cloud MCP connectors alongside `managed-mcp.json`
+- Fixed a PowerShell permission bypass: built-in `cd` functions (`cd..`, `cd\`, `cd~`, `X:`) changed the working directory undetected, letting a later command read outside the workspace
+- Fixed the sandbox write allowlist in git worktrees covering the entire main repository root instead of only the shared `.git` directory (with `hooks/` and `config` denied)
+- Fixed PowerShell prefix/wildcard allow rules (e.g. `PowerShell(dotnet.exe build *)`) not pre-approving native executables and scripts
+- Fixed a permission-analysis gap where the parser trusted stale variable-tracking values for `PWD`/`OLDPWD`/`DIRSTACK` across `cd`/`pushd`/`popd`
+- Fixed `find` in the Bash tool exhausting the macOS system file/vnode table and crashing the host on large directory trees
+- Fixed the managed-settings approval dialog leaving the terminal frozen after accepting at startup
+- Fixed `/ultraplan` and remote session creation failing with "Could not capture uncommitted changes" when the working tree has no real changes
+- Fixed `otelHeadersHelper` failing silently when the script path contains spaces; helper failures are now reported in `/doctor` and the debug log
+- Fixed the thinking spinner staying amber across tool calls and onto fresh thinking bursts
+- Fixed collapsed Bash output reporting the wrong hidden-line count for outputs with many short lines
+- Fixed slash-command argument-hint clipping trailing typed characters when the hint overflows the input box
+- Fixed argument-hint and progressive arg suggestions not appearing after Tab-completing a skill whose frontmatter `name:` differs from its directory basename
+- Fixed the status bar showing the user's baseline `/effort` setting instead of the effort level applied by skill/agent `effort:` frontmatter
+- Fixed Ctrl+O transcript view freezing at the moment it was opened instead of tailing new messages
+- Fixed editing a recalled prompt-history entry losing the edit when navigating further up/down with arrow keys
+- Fixed `/config` exit summary reporting phantom changes to auto-compact and theme when toggling unrelated settings
+- Fixed `/insights` crashing when cached session-meta files are missing optional fields
+- Fixed malformed PowerShell and History tool calls with missing input being misclassified as reads in transcript collapsing
+- Fixed renaming a Remote Control session from claude.ai or the Claude mobile app not updating the local session name for `claude --resume`
+- Fixed a race where a just-submitted prompt could appear twice in the up-arrow history
+- Fixed tapping the "Jump to bottom" pill in fullscreen mode not dismissing it immediately
+- Improved `/feedback` reports to include the conversation that happened before context compaction, making issues from earlier in long sessions easier to triage
+
+## 2.1.148
+
+- Fixed the Bash tool returning exit code 127 on every command for some users (a regression introduced in 2.1.147)
+
 ## 2.1.147
 
 - Pinned background sessions (`Ctrl+T` in `claude agents`) now stay alive when idle, are restarted in place to apply Claude Code updates, and are shed under memory pressure only after non-pinned sessions
 - Renamed `/simplify` to `/code-review`. It now reports correctness bugs at a chosen effort level (e.g., `/code-review high`); pass `--comment` to post findings as inline GitHub PR comments. The old cleanup-and-fix behavior has been removed
-- Hardened the REPL tool sandbox against prototype-pollution and thenable-based escapes
 - Improved auto-updater: retries transient network failures, reports specific error categories and OS error codes on failure, and shows the current version when an update fails
 - Improved diff rendering performance for large file edits
 - Prompt history no longer records consecutive duplicate entries — recalling a prompt with arrow-up and submitting it again won't add another copy

feed.xml

@@ -6,16 +6,60 @@
   <author><name>Anthropic</name></author>
   <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md"/>
   <link rel="self" type="application/atom+xml" href="https://raw.githubusercontent.com/anthropics/claude-code/main/feed.xml"/>
-  <updated>2026-05-21T20:39:15Z</updated>
+  <updated>2026-05-23T04:03:45Z</updated>
+  <entry>
+    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.150</id>
+    <title>Claude Code v2.1.150</title>
+    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.150"/>
+    <updated>2026-05-23T04:03:45Z</updated>
+    <content type="html">&lt;p&gt;• Internal infrastructure improvements (no user-facing changes)&lt;/p&gt;</content>
+  </entry>
+  <entry>
+    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.149</id>
+    <title>Claude Code v2.1.149</title>
+    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.149"/>
+    <updated>2026-05-22T22:09:22Z</updated>
+    <content type="html">&lt;p&gt;• /usage now shows a per-category breakdown of what's driving your limits usage — skills, subagents, plugins, and per-MCP-server cost&lt;/p&gt;
+&lt;p&gt;• /diff detail view can now be scrolled with the keyboard (arrows, j/k, PgUp/PgDn, Space, Home/End)&lt;/p&gt;
+&lt;p&gt;• Markdown output now renders GFM task list checkboxes (- [ ] todo / - [x] done) instead of plain bullets&lt;/p&gt;
+&lt;p&gt;• Enterprise: added the allowAllClaudeAiMcps managed setting to load claude.ai cloud MCP connectors alongside managed-mcp.json&lt;/p&gt;
+&lt;p&gt;• Fixed a PowerShell permission bypass: built-in cd functions (cd.., cd\, cd~, X:) changed the working directory undetected, letting a later command read outside the workspace&lt;/p&gt;
+&lt;p&gt;• Fixed the sandbox write allowlist in git worktrees covering the entire main repository root instead of only the shared .git directory (with hooks/ and config denied)&lt;/p&gt;
+&lt;p&gt;• Fixed PowerShell prefix/wildcard allow rules (e.g. PowerShell(dotnet.exe build *)) not pre-approving native executables and scripts&lt;/p&gt;
+&lt;p&gt;• Fixed a permission-analysis gap where the parser trusted stale variable-tracking values for PWD/OLDPWD/DIRSTACK across cd/pushd/popd&lt;/p&gt;
+&lt;p&gt;• Fixed find in the Bash tool exhausting the macOS system file/vnode table and crashing the host on large directory trees&lt;/p&gt;
+&lt;p&gt;• Fixed the managed-settings approval dialog leaving the terminal frozen after accepting at startup&lt;/p&gt;
+&lt;p&gt;• Fixed /ultraplan and remote session creation failing with "Could not capture uncommitted changes" when the working tree has no real changes&lt;/p&gt;
+&lt;p&gt;• Fixed otelHeadersHelper failing silently when the script path contains spaces; helper failures are now reported in /doctor and the debug log&lt;/p&gt;
+&lt;p&gt;• Fixed the thinking spinner staying amber across tool calls and onto fresh thinking bursts&lt;/p&gt;
+&lt;p&gt;• Fixed collapsed Bash output reporting the wrong hidden-line count for outputs with many short lines&lt;/p&gt;
+&lt;p&gt;• Fixed slash-command argument-hint clipping trailing typed characters when the hint overflows the input box&lt;/p&gt;
+&lt;p&gt;• Fixed argument-hint and progressive arg suggestions not appearing after Tab-completing a skill whose frontmatter name: differs from its directory basename&lt;/p&gt;
+&lt;p&gt;• Fixed the status bar showing the user's baseline /effort setting instead of the effort level applied by skill/agent effort: frontmatter&lt;/p&gt;
+&lt;p&gt;• Fixed Ctrl+O transcript view freezing at the moment it was opened instead of tailing new messages&lt;/p&gt;
+&lt;p&gt;• Fixed editing a recalled prompt-history entry losing the edit when navigating further up/down with arrow keys&lt;/p&gt;
+&lt;p&gt;• Fixed /config exit summary reporting phantom changes to auto-compact and theme when toggling unrelated settings&lt;/p&gt;
+&lt;p&gt;• Fixed /insights crashing when cached session-meta files are missing optional fields&lt;/p&gt;
+&lt;p&gt;• Fixed malformed PowerShell and History tool calls with missing input being misclassified as reads in transcript collapsing&lt;/p&gt;
+&lt;p&gt;• Fixed renaming a Remote Control session from claude.ai or the Claude mobile app not updating the local session name for claude --resume&lt;/p&gt;
+&lt;p&gt;• Fixed a race where a just-submitted prompt could appear twice in the up-arrow history&lt;/p&gt;
+&lt;p&gt;• Fixed tapping the "Jump to bottom" pill in fullscreen mode not dismissing it immediately&lt;/p&gt;
+&lt;p&gt;• Improved /feedback reports to include the conversation that happened before context compaction, making issues from earlier in long sessions easier to triage&lt;/p&gt;</content>
+  </entry>
+  <entry>
+    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.148</id>
+    <title>Claude Code v2.1.148</title>
+    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.148"/>
+    <updated>2026-05-22T01:16:46Z</updated>
+    <content type="html">&lt;p&gt;• Fixed the Bash tool returning exit code 127 on every command for some users (a regression introduced in 2.1.147)&lt;/p&gt;</content>
+  </entry>
   <entry>
     <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.147</id>
     <title>Claude Code v2.1.147</title>
     <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.147"/>
     <updated>2026-05-21T20:39:15Z</updated>
-    <content type="html">&lt;p&gt;• Added the Workflow tool for deterministic multi-agent orchestration. It is off by default — set CLAUDE_CODE_WORKFLOWS=1 to enable&lt;/p&gt;
-&lt;p&gt;• Pinned background sessions (Ctrl+T in claude agents) now stay alive when idle, are restarted in place to apply Claude Code updates, and are shed under memory pressure only after non-pinned sessions&lt;/p&gt;
+    <content type="html">&lt;p&gt;• Pinned background sessions (Ctrl+T in claude agents) now stay alive when idle, are restarted in place to apply Claude Code updates, and are shed under memory pressure only after non-pinned sessions&lt;/p&gt;
 &lt;p&gt;• Renamed /simplify to /code-review. It now reports correctness bugs at a chosen effort level (e.g., /code-review high); pass --comment to post findings as inline GitHub PR comments. The old cleanup-and-fix behavior has been removed&lt;/p&gt;
-&lt;p&gt;• Hardened REPL and Workflow tool sandboxes against prototype-pollution and thenable-based escapes&lt;/p&gt;
 &lt;p&gt;• Improved auto-updater: retries transient network failures, reports specific error categories and OS error codes on failure, and shows the current version when an update fails&lt;/p&gt;
 &lt;p&gt;• Improved diff rendering performance for large file edits&lt;/p&gt;
 &lt;p&gt;• Prompt history no longer records consecutive duplicate entries — recalling a prompt with arrow-up and submitting it again won't add another copy&lt;/p&gt;
@@ -592,80 +636,4 @@
 &lt;p&gt;• PowerShell tool: bare -- (e.g. git diff -- file) is no longer mis-flagged as the --% stop-parsing token&lt;/p&gt;
 &lt;p&gt;• Fixed Agent SDK hang when the model emits a malformed tool name in a parallel tool call batch&lt;/p&gt;</content>
   </entry>
-  <entry>
-    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.123</id>
-    <title>Claude Code v2.1.123</title>
-    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.123"/>
-    <updated>2026-05-18T01:52:01Z</updated>
-    <content type="html">&lt;p&gt;• Fixed OAuth authentication failing with a 401 retry loop when CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS=1 is set&lt;/p&gt;</content>
-  </entry>
-  <entry>
-    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.122</id>
-    <title>Claude Code v2.1.122</title>
-    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.122"/>
-    <updated>2026-05-18T01:52:01Z</updated>
-    <content type="html">&lt;p&gt;• Added ANTHROPIC_BEDROCK_SERVICE_TIER environment variable to select a Bedrock service tier (default, flex, or priority), sent as the X-Amzn-Bedrock-Service-Tier header&lt;/p&gt;
-&lt;p&gt;• Pasting a PR URL into the /resume search box now finds the session that created that PR (GitHub, GitHub Enterprise, GitLab, and Bitbucket)&lt;/p&gt;
-&lt;p&gt;• /mcp now shows claude.ai connectors hidden by a manually-added server with the same URL, with a hint to remove the duplicate&lt;/p&gt;
-&lt;p&gt;• Clarified the /mcp message shown when an MCP server is still unauthorized after the browser sign-in flow&lt;/p&gt;
-&lt;p&gt;• OpenTelemetry: numeric attributes on api_request/api_error log events are now emitted as numbers, not strings&lt;/p&gt;
-&lt;p&gt;• OpenTelemetry: added claude_code.at_mention log event for @-mention resolution&lt;/p&gt;
-&lt;p&gt;• Fixed /branch producing forks that fail with "tool_use ids were found without tool_result blocks" when the source session contained entries from rewound timelines&lt;/p&gt;
-&lt;p&gt;• Fixed /model not showing the Effort option for Bedrock application inference profile ARNs, and those ARNs not receiving output_config.effort&lt;/p&gt;
-&lt;p&gt;• Fixed Vertex AI / Bedrock returning invalid_request_error: output_config: Extra inputs are not permitted on session-title generation and other structured-output queries&lt;/p&gt;
-&lt;p&gt;• Fixed Vertex AI count_tokens endpoint returning 400 errors for users behind proxy gateways&lt;/p&gt;
-&lt;p&gt;• Fixed spinnerTipsOverride.excludeDefault not suppressing the time-based spinner tips&lt;/p&gt;
-&lt;p&gt;• Fixed ToolSearch missing MCP tools that connected after session start in nonblocking mode&lt;/p&gt;
-&lt;p&gt;• Fixed !exit / !quit in bash mode terminating the CLI instead of running as a shell command&lt;/p&gt;
-&lt;p&gt;• Fixed images sent to newer models being resized to 2576px per side instead of the correct 2000px maximum&lt;/p&gt;
-&lt;p&gt;• Fixed remote control session idle status redrawing twice per second, which could flood tmux -CC control pipes and pause the terminal&lt;/p&gt;
-&lt;p&gt;• Fixed assistant messages appearing blank in some sessions due to a stale view preference&lt;/p&gt;
-&lt;p&gt;• Fixed a malformed hooks entry in settings.json no longer invalidating the entire file&lt;/p&gt;
-&lt;p&gt;• Voice mode: keybindings bound to Caps Lock now show an error since terminals don't deliver Caps Lock as a key event&lt;/p&gt;</content>
-  </entry>
-  <entry>
-    <id>https://github.com/anthropics/claude-code/releases/tag/v2.1.121</id>
-    <title>Claude Code v2.1.121</title>
-    <link rel="alternate" type="text/html" href="https://github.com/anthropics/claude-code/releases/tag/v2.1.121"/>
-    <updated>2026-05-18T01:52:01Z</updated>
-    <content type="html">&lt;p&gt;• Added alwaysLoad option to MCP server config — when true, all tools from that server skip tool-search deferral and are always available&lt;/p&gt;
-&lt;p&gt;• Added claude plugin prune to remove orphaned auto-installed plugin dependencies; plugin uninstall --prune cascades&lt;/p&gt;
-&lt;p&gt;• Added a type-to-filter search box to /skills so you can find a skill in long lists without scrolling&lt;/p&gt;
-&lt;p&gt;• PostToolUse hooks can now replace tool output for all tools via hookSpecificOutput.updatedToolOutput (previously MCP-only)&lt;/p&gt;
-&lt;p&gt;• Fullscreen mode: typing into the prompt no longer jumps scroll back to the bottom after you've scrolled up to read earlier output&lt;/p&gt;
-&lt;p&gt;• Dialogs that overflow the terminal are now scrollable with arrow keys, PgUp/PgDn, home/end, and mouse wheel in both fullscreen and non-fullscreen modes&lt;/p&gt;
-&lt;p&gt;• Clicking any line of a long URL that wraps across rows in fullscreen mode now opens the full URL&lt;/p&gt;
-&lt;p&gt;• SDK and claude -p: CLAUDE_CODE_FORK_SUBAGENT=1 now works in non-interactive sessions&lt;/p&gt;
-&lt;p&gt;• --dangerously-skip-permissions no longer prompts for writes to .claude/skills/, .claude/agents/, and .claude/commands/&lt;/p&gt;
-&lt;p&gt;• /terminal-setup now enables iTerm2's "Applications in terminal may access clipboard" setting so /copy works, including from tmux&lt;/p&gt;
-&lt;p&gt;• MCP servers that hit a transient error during startup now auto-retry up to 3 times instead of staying disconnected&lt;/p&gt;
-&lt;p&gt;• The terminal tab session title is now generated in your configured language setting&lt;/p&gt;
-&lt;p&gt;• Claude.ai connectors with the same upstream URL are now deduplicated instead of appearing as duplicates&lt;/p&gt;
-&lt;p&gt;• Vertex AI: support X.509 certificate-based Workload Identity Federation (mTLS ADC)&lt;/p&gt;
-&lt;p&gt;• Faster startup after upgrading: removed the Recent Activity panel from the release-notes splash&lt;/p&gt;
-&lt;p&gt;• LSP diagnostic summaries now expand on click/ctrl+o and show the expand hint&lt;/p&gt;
-&lt;p&gt;• SDK: mcp_authenticate now supports redirectUri for custom scheme completion and claude.ai connectors&lt;/p&gt;
-&lt;p&gt;• OpenTelemetry: added stop_reason, gen_ai.response.finish_reasons, and user_system_prompt (gated behind OTEL_LOG_USER_PROMPTS) to LLM request spans&lt;/p&gt;
-&lt;p&gt;• [VSCode] Voice dictation now respects the accessibility.voice.speechLanguage setting when no Claude Code language is configured&lt;/p&gt;
-&lt;p&gt;• [VSCode] /context now opens a native token usage dialog&lt;/p&gt;
-&lt;p&gt;• Fixed unbounded memory growth (multi-GB RSS) when processing many images in a session&lt;/p&gt;
-&lt;p&gt;• Fixed /usage leaking up to ~2GB of memory on machines with large transcript histories&lt;/p&gt;
-&lt;p&gt;• Fixed memory leak when long-running tools fail to emit a clear progress event&lt;/p&gt;
-&lt;p&gt;• Fixed Bash tool becoming permanently unusable when the directory Claude was started in is deleted or moved mid-session&lt;/p&gt;
-&lt;p&gt;• Fixed --resume crashing on startup in external builds&lt;/p&gt;
-&lt;p&gt;• Fixed --resume failing on large sessions when a transcript line was corrupted by an unclean shutdown — the corrupt line is now skipped&lt;/p&gt;
-&lt;p&gt;• Fixed thinking.type.enabled is not supported error when using Bedrock application inference profile ARNs&lt;/p&gt;
-&lt;p&gt;• Fixed Microsoft 365 MCP OAuth failing with duplicate or unsupported prompt parameter&lt;/p&gt;
-&lt;p&gt;• Fixed scrollback duplication when pressing Ctrl+L or triggering a redraw in non-fullscreen mode on tmux, GNOME Terminal, Windows Terminal, and Konsole&lt;/p&gt;
-&lt;p&gt;• Fixed claude.ai MCP connectors silently disappearing when the connector-list fetch hits a transient auth error at startup&lt;/p&gt;
-&lt;p&gt;• Fixed "Always allow" rules for built-in tools in remote sessions not surviving worker restarts&lt;/p&gt;
-&lt;p&gt;• Fixed NO_PROXY not being respected for all HTTP clients when set via managed-settings.json under the native build&lt;/p&gt;
-&lt;p&gt;• Fixed managed settings approval prompt exiting the session even when accepted — now applies settings and continues&lt;/p&gt;
-&lt;p&gt;• Fixed /usage returning "rate limited" after a stale OAuth token — now refreshes automatically&lt;/p&gt;
-&lt;p&gt;• Fixed invalid legacy enum values in settings.json invalidating the entire settings file&lt;/p&gt;
-&lt;p&gt;• Fixed /usage dialog content being clipped when no-flicker mode is off&lt;/p&gt;
-&lt;p&gt;• Fixed /focus showing "Unknown command" when the fullscreen renderer is off — now explains how to enable it&lt;/p&gt;
-&lt;p&gt;• Fixed embedded grep/find/rg shell wrappers failing when the running binary is deleted mid-session — now falls back to installed tools&lt;/p&gt;
-&lt;p&gt;• Reduced peak file descriptor usage during find in the Bash tool on large directory trees&lt;/p&gt;</content>
-  </entry>
 </feed>