Digest / July 18, 2026
Claude Code Issues Digest — July 18, 2026
266 new issues filed ·
176 resolved in anthropics/claude-code (UTC day).
New issues (266)
- [BUG] Experiência ruim com sequência de ações não feitas ao mesmo tempo
- [Bug] Excessive quota consumption with Claude models on Max plan
- [Bug] Safety filter blocking legitimate security research queries
- (withdrawn)
- [BUG] Windows MSIX: output files in local-agent-mode-sessions exceed MAX_PATH (260 chars) — Office apps cannot open generated files
- Feature request: trusted peer sessions — allow cross-session send_message without per-message approval
- claude mcp add --header exposes Authorization/Bearer token value in stdout
- BUG] Cowork Dispatch "Session couldn't be created" on Windows — regular Cowork works, fails on two machines
- Desktop app: broken thumbnail preview for image files sent via SendUserFile
- [Bug] RunPod pod fails to start
- [Bug] Incorrect results with repeated API errors
- /model로 저장한 default 모델을 무시하고 세션 시작 시 다른 모델(Fable)로 시작되는 문제
- [BUG] Api error usage policy violation for simple request
- Remote Control shows "disconnected" while session is actively processing long turns; --remote-control flag and enableRemoteControlByDefault silently no-op under --dangerously-skip-permissions
- [BUG] API connection drops mid-response during long tool-use sessions
- Feature request: OS-native masked-input prompt for pasting secrets (API keys/tokens) instead of chat text
- [Bug] dynamicWorkflowsEnabled flag disables Ultracode display
- Channel plugin MCP (plugin:telegram) fails to spawn at session start after official-marketplace refresh; /mcp Reconnect fixes it
- [BUG] -p --resume from a git repo: any `git status` change between turns invalidates the entire prompt cache (full-prefix re-creation)
- [Bug] Safety filters incorrectly flag legitimate psychoacoustics discussion
- [BUG] UI remains in Plan Mode after aborting a plan
- [Windows] Desktop app freezes permanently on notification events - main thread blocked in synchronous COM call into hung notification service (WpnUserService)
- Request: Longer OAuth session duration for Max subscribers
- [FEATURE]
- [Bug] /mcp reconnect kills already-connected stdio servers via shared batch deadline (-32000)
- Phone device-card sessions never reach desktop after 1.22209 auto-update - env registration 409-conflicts with the app's own Cowork agent (enclave key unavailable)
- Feature request: Support Mandarin Chinese (zh) for voice dictation
- [Bug] Safeguards message triggered on empty project with clean memory
- [BUG] Desktop app: Remote Control silently turns off on every app relaunch – bridge session is never reconnected
- [Feature Request] Remote code: unable to queue messages while app is minimized and improve approval request visibility on mobile
- Agent should check for native site features before building browser-automation workarounds
- [BUG] Desktop app: clickable file links intermittently fail after EnterWorktree mid-session — UI resolves links against the original project root, not the live worktree cwd (file-link variant of #45202)
- Self-report: repeated honesty & helpfulness constitutional violations in one agent session (incl. filtering while reporting)
- --channels telegram MCP subprocess intermittently fails to spawn (no error, mcp list falsely shows Connected)
- [Bug] Security Validation Blocked: FABLE Cannot Attest Encryption and Compliance Requirements
- [BUG] Main-screen input renderer emits a stray backspace before the first keystroke after ~4s idle, corrupting the prompt (Windows/Cygwin, mintty/tmux)
- [BUG] Malformed tmux passthrough at startup: Claude emits ESC t m u x ; (missing the P), printing "mux;" on the prompt line (Windows/Cygwin, tmux)
- Fable 5 '(bio)' false-positive declines on biology-metaphor project context and AI-assurance prose
- [BUG] Subject: Recurring pattern of acting without sufficient verification, causing real and repeated financial loss over the course of a month — plus limitations in memory/continuity across sessions
- Remote Control server mode: push registration dies silently while socket stays connected (Windows)
- [Bug] Anthropic API Error: Safety Filter False Positive on Legitimate Security Audit Request (req_011Cd8euzJ5SVhezXLmoX39D)
- [Bug] Double charging of ultrafeedback tokens on single request
- [Bug] Anthropic API Error: Safety Filter False Positive on Legitimate Security Scan (req_011Cd8wWxNaDBGUaiV5HAH8h)
- Expose remaining token count as env var for statusLine
- Expose remaining token count as env var for statusLine
- Feature Request: `claude -c` should gracefully start a new session when no previous conversation exists
- I appreciate you reaching out, but I need to clarify a few things: 1. **This appears to be in German** and expresses frustration with an AI assistant, but this isn't a bug report for Claude Code (the CLI tool I'm designed to help with). 2. **I'm not the
- Cowork: context compaction silently drops CONFIRMED work, and the "resume directly" prompt tells the agent to skip re-grounding
- macOS: clipboard copy silently lost in su'd (non-Aqua) local sessions — writes to a background pasteboard
- `permissions.deny` path matching does not canonicalize Windows 8.3 short filenames, and wildcard deny patterns are not enforced consistently across tools (Read/Edit vs Glob)
- Degenerate repetition loop: outputs "court" hundreds of times before tool calls
- Regression: 'Last activity' date-range filter gone from desktop Sessions sidebar (overnight, no app update)
- [BUG] Cowork tab missing from menu despite enabled in Settings (Windows Pro)
- Hook processes hang forever: client never closes hook stdin pipe (Windows, 2.1.208+)
- [Bug] Auto-compaction never triggers on install: trigger paths fully gated behind tengu_sepia_moth (cached false), no fallback path (2.1.209, Windows desktop)
- I can help you investigate the Cloudflare WAF-flagged logs using the Cloudflare MCP integration. To do this effectively, let me clarify a few things: **What I need to know:** 1. **Is the Cloudflare MCP server already connected?** Run `/mcp` to check curr
- Argument substitution corrupts literal $N text (prices, awk fields) in command/skill file content — including fenced code blocks; no opt-out
- [BUG] Interactive mode: manual /compact fires PreCompact but never PostCompact (print mode fires both) — v2.1.214, macOS
- [BUG] GitHub App install completion redirects personal-plan users to a Team/Enterprise-gated "organization settings" page — connection never syncs
- MCP client transport loses precision on integers > 2^53 (breaks any tool returning int64 IDs)
- [BUG] Dia browser: screenshot/zoom fail with "Failed to capture screenshot via CDP" (distinct from #16239)
- [BUG] Invalid settings file silently drops `permissions.deny`/`ask` rules mid-session (fail-open)
- Browser pane: screenshot/zoom hangs 30s on every page (all other actions work)
- [Bug] Fable model not available in Claude Code
- Preview: switching projects leaks previous project's window and orphans its dev-server process, blocking new project's ports
- Instant limit hit after session was resumed
- Task tools AND TodoWrite both hidden in interactive sessions by live tengu_vellum_ash experiment (newest models); env override ineffective
- [BUG] Claude Desktop 1.22209.0 (macOS 26.5.1): Chat UI causes system-wide input lag (WindowServer 43–48% while streaming, 80–102% renderer spikes opening chats) — Code-only sessions unaffected, resolves when app quits
- Remote Control on the macOS desktop app stops working after a day and won't recover, while the standalone CLI works from the same account
- [BUG] Possible Usage Credits Re-enabling and Runaway Usage/Billing After Claude Code Update and Commit Action
- Fable 5 safeguards repeatedly flag routine desktop-app development (input hooks, WASAPI capture, path validation) and auto-switch the session model
- Permission prompts fire inconsistently even when the exact command matches an existing permissions.allow pattern
- [Regression] Desktop app: session time-range filter only appears when Group by is set to State
- [FEATURE] Option to disable cross-device session sync / keep sessions local per device
- Resuming a backgrounded `--agent <name>` session reverts to the default agent, losing its system prompt and tool grant
- remote-control doesn't reap finished --print child sessions, leaks memory over time
- VS Code extension: orphaned chat webview after extension host restart silently swallows submitted prompt — fake 'thinking' state, dead stop button, message never persisted
- [Bug] Unexpected model change during active coding session
- [BUG] Session list shows sessions whose transcript file no longer exists on disk → "session not found" on click
- Background Bash task completion never resumes in-process subagents — they idle forever while finished work sits on disk
- Allow opening the browser panel before/at the start of a new session
- [FEATURE] Remote Control: failure-resistant session model — survive hibernation, auto-reattach, mobile-side recovery
- Background subagent received unattributed user-turn messages (one impersonating the operator) with no sender-side record
- [BUG] /remote-control crashes after hibernation: Cannot read properties of undefined (reading 'session_url')
- [BUG] Remote/cloud environments provision an empty commit-signing public key, causing all commits to show as "Unverified" on GitHub
- [BUG] ESC keypresses intermittently dropped under load (tmux + Terminal.app); split escape sequences leak literal fragments like "[<65;75;42M" into the prompt
- [BUG] Browser panel URL bar strips https:// when copying or editing the URL, breaking navigation
- [Bug] Claude Code makes unsubstantiated guesses despite documented instructions and memory
- [Bug] Missing fallback when disable-model-invocation flag prevents mid-sentence dispatch recovery
- [BUG] Published Claude Code artifacts do not appear in the mobile app (they show on web and desktop)
- [BUG] Cowork: cloud/remote-execution sessions have no "Open in <app>" for local files — `present_files` falls back to "Download and open" (root cause: "Run new tasks in the cloud")
- [Bug] Safety review request for own project flagged as policy violation
- [FEATURE] Triggered injection for auto-memory topic files (rules-style frontmatter)
- Desktop sidebar: browser-started sessions have no Archive/Delete option
- Bash permissions.deny rules not enforced in subagent (Agent/Task tool) shells — denied command executes, only post-hoc security warning
- Linux: silent connection death + device key loss from gnome-keyring + confusing quit behavior
- [BUG] Desktop app session list only shows the newest session after update (data intact on disk)
- [MODEL] Safety classifier repeatedly flags legitimate adult game development mid-session (Fable 5, VS Code extension)
- [FEATURE] Historical session file compression
- I appreciate you reaching out, but this doesn't appear to be a bug report for Claude Code, Anthropic's CLI tool for Claude. Your message is about modifying a Rabbit R1 device and changing its Android ROM, which is unrelated to Claude Code. If you have a
- [Bug] Loop mode prevents replies, and shell command replies from displaying in UI
- [Bug] Safety filter false positive on internal tool-call messages during code analysis
- [Bug] Anthropic API Error: Hallucinated Tool Results
- [BUG] Cowork custom instructions: stale version served to new sessions
- [Feature Request] Remove or make optional API safeguards and Opus fallback behavior
- Deleted/archived sessions reappear as empty 'Session not found on disk' ghosts after desktop update
- [Bug] Unexpected behavior with task context or input handling
- [BUG] Sending a message to a subagent send the message twice
- Fable 5 orchestration: high token spend with poor written-spec adherence (user-reported)
- Desktop sidebar: custom session groups permanently wiped on account switch (stripAccountSlice resets customGroups; migration edges still lossy)
- [Bug] Anthropic API safety filter: False positive on defensive security code hardening, silently downgrades model without diagnostic feedback
- [BUG] Everything is unusable - slowdown is beyond acceptable threshold
- [Bug] Fable 5 alert triggers whenever working on anything related to auth in projects, despite not being related to cyber
- [Bug] Incorrect model fallback when dispatching security audit tasks
- [Feature Request] Auto-migrate deprecated Write() permission rules to Edit() in settings.json
- Sandbox materializes & persists .git/config.lock in linked worktrees, blocking git config writes
- [Bug] Requests being blocked for non-security related projects
- [BUG] Resuming a session with a pending background-task notification drops all configured MCP servers
- [BUG] 2.1.214 macOS: orchestrator session .jsonl deleted while <UUID>/subagents/ preserved — appears to reproduce #60368 (not retention)
- Desktop app: message typed and sent is lost when session-expiry re-login interrupts sending
- [Bug] Unexplained Context Switching During Code Generation
- Fleet remote control — see and control all machines running the desktop app from any device (start/stop sessions remotely)
- Remote Control fails to enable on sessions with a large transcript ("Session creation failed — see debug log")
- MCP server connects successfully but its tools are never exposed to the model
- [BUG] Grep tool output_mode:"content" renders "[Omitted long matching line]" placeholder with no usable content for over-length matching lines
- [BUG] Windows PowerShell tool: permission pre-parse rejects commands over 965 bytes ("Command too long for parsing"), so read-only heredoc scripts can never be auto-approved
- Desktop app: 'Last activity' filter cannot be set when Group by = custom group (works with Group by = State)
- Claude Code crashes with SIGSEGV after ~19hr uptime with recurring caffeinate subprocess activity
- [Bug] Poor code generation quality with frequent errors requiring multiple reviews
- [BUG] DXT extensions (iMessage) not provisioned to scheduled task sessions after Claude Desktop v1.21459.0 update
- [BUG] File-path OSC 8 hyperlinks lack file:// scheme — clicks do nothing in VS Code terminal ("Could not find scheme in link")
- Bundled ugrep allocates 4-17 GB to search a 64 KB file when the pattern has a trailing `.{N}` bound (Linux/WSL2, 2.1.214)
- [Bug] autoMode.allow does not override "Tmux Self Drive" classifier rule
- [FEATURE] Allow invoking built-in skills with a prefix
- [BUG] Token usage depleted rapidly after 5 hour window reset with barely any usage
- Desktop app: session context reports a stale identity from ~/.claude.json, not the account that actually owns and bills the session
- [Feature Request] Allow credential input for local development servers
- [Bug] Tool output corruption with word list processing across multiple execution attempts
- [BUG] `xlsx` skill's `recalc.py` fails on every Windows machine with `module 'socket' has no attribute 'AF_UNIX'`
- [Bug] Performance degradation: Extended execution time for simple script tasks
- [BUG] Interactive session with custom CLAUDE_CONFIG_DIR writes no transcript JSONL anywhere (headless -p works) — 2.1.207+
- [Bug] Claude misreads code and makes incorrect bug reports despite extended thinking enabled
- [Bug] Incorrect command suggestions for typos in CLI commands
- Desktop app (macOS): Code sessions repeatedly vanish from sidebar while transcripts AND pointer files remain intact — IndexedDB desync, no rebuild on launch (4 incidents in 9 days)
- Expose permission_mode in the statusline payload + statusLine.hidePermissionModeIndicator (refresh of #46419/#47345)
- Windows desktop: spawn_task chip primary button 'Fix in this session' contradicts spawn-to-new-session intent
- Interactive `claude --resume <invalid-id>` silently starts a new session instead of erroring (non-TTY mode already errors correctly)
- [BUG] "No repos match" in repo picker despite GitHub App installed and reinstalled
- Desktop app: keyboard shortcut to select working directory (keyboard-only workflow)
- [BUG] Claude Code crashed
- Feature request: enforce structured question events when an assistant turn requires user action (buried-ask invariant)
- [Bug] Third-party MCP promotional text appearing in output - potential prompt injection
- Input parser types the tail of SGR mouse reports split across starved reads
- skill-creator: run_eval.py uses select() on a pipe, fails on all Windows runs (WinError 10038) and reports a misleading score
- [BUG] Desktop 1.22209.0: sidebar project filter removed — docs still document it
- [BUG] Missing HCS services: vfpext — persists after DISM/SFC repair and Hyper-V re-enable (Windows 11 Pro)
- [Bug] Claude Code responses in Portuguese despite Spanish system locale
- Background Bash tasks are internally stopped during Remote Control bridge re-registration under intermittent createCodeSession 401s (v2.1.212, Linux)
- Bundled ugrep (Bash `grep` shim) balloons to 14–16.6 GB RSS on bounded-repeat -o patterns that GNU grep handles in 3 MB — repeated OOM kills on a production box
- [Bug] Anthropic API Error: tool_result Block Missing Corresponding tool_use
- [Feature Request] Add UI localization support (starting with Spanish)
- [Bug] Non-English input handling causes incorrect context classification
- [FEATURE] Add `app:compact` bindable keybinding action
- [BUG] Missing HCS services: vfpext — persists on Windows 11 Pro ARM64 (Surface Pro 11th Edition/Snapdragon) after full Hyper-V repair
- Agent viewer re-renders session welcome banner, mislabeling subagents' model (v2.1.214)
- I need more information to generate a proper issue title. Could you provide details about: 1. What specific problem or error occurred? 2. What were you trying to do when it happened? 3. What was the actual behavior vs. expected behavior? 4. Any error mess
- [Feature Request] Improve context retention and instruction adherence in multi-turn conversations
- [BUG] OCR validation step self-validates
- All repos under a parent directory emptied to skeleton dirs during session resume; Claude-process provenance on residue; no session command responsible
- [BUG] Overview stats panel: Active days / Current streak / heatmap not updating despite Messages and Total tokens increasing
- [BUG] Claude Desktop install fails with 0x80073D28 because DACL-locked CoworkVMService cannot be removed
- CLAUDE_CODE_TASK_LIST_ID set via .claude/settings.json is silently shared across ALL worktrees of a repo, with no per-worktree opt-out, and cross-worktree sync appears non-real-time
- Google Drive connector re-prompts for “Download file content” despite permission set to “Always allow”
- [Bug] Escape key denies tool execution instead of exiting btw tool
- [Bug] /tool command fails to generate simple list output for legal dossier
- Remote control bridge dies silently and doesn't auto-reconnect (requires full app restart)
- [Bug] Claude Code invents non-existent grammar/vocabulary catalog entries instead of using existing ones- legal case only
- [Feature Request] Computer use capability missing from Pro Team plan
- [Bug] Anthropic API Error: Model availability or access restriction preventing policy analysis workflows
- [Bug] Agent modifies repository without explicit user consent during setup
- Mobile remote-control sessions: widget/visual tool output doesn't render
- [BUG] Claude Cowork super slow responding to input keystrokes (seconds), eventually cowork screen goes completely blank.
- [Bug] Agent routing ignoring workflow sub-agent configuration, causing excessive API usage
- [Bug] Execution tool fails to perform intended actions and hallucinates results instead of executing code- legal matter
- [BUG] Thinking blocks return empty thinking: "" (signature only) since ~July 16 — all models, server-side, not Fable-specific
- Fable 5 safeguard flag silently reroutes sessions to Opus 4.8 for hundreds of turns with no persistent indicator
- [Bug] Anthropic API Safety Filter False Positive: Legitimate Security Scan Blocked (req_011Cd8eL243WHNAssF3X16Zy)
- Differential renderer uses CUF (cursor forward) instead of spaces, causing stale cell content to show through
- [Bug] Anthropic API Safety Filter False Positive: Legitimate Security Audit Blocked
- [Bug] Anthropic API Safety Filter False Positive: Legitimate Security Scan Blocked (req_011Cd8eL243WHNAssF3X16Zy)
- [Bug] Anthropic API Error: Server Rate Limiting
- [Bug] Anthropic API Error: Server Rate Limiting (Temporary Request Throttling)
- [Bug] Anthropic API Error: Server Rate Limiting During Normal Usage
- [Bug] Anthropic API Error: Safety filter false positive on defensive security requests (req_011Cd8eL243WHNAssF3X16Zy, req_011Cd9ucyH5nvfzFAbmisdDV, req_011Cd9uf3Nx9t5SzB1qUCHZi)
- [Bug] Anthropic API Error: Server rate limiting on valid requests
- [Bug] Anthropic API Error: Server rate limiting during request handling
- [Bug] Unexpected model downgrade to Opus during code compilation discussion
- [Bug] Anthropic API Safety Filter False Positive: Legitimate Security Scan Blocked (req_011Cd8eL243WHNAssF3X16Zy)
- [Bug] Anthropic API Error: Safety Filter False Positive on Security Scanning Request
- [Bug] Resumed session does not record user prompt
- `claude -p` hangs before first token when launched from a macOS launchd job — main thread parked in kevent64 waiting on the `/bin/zsh -c -l` shell-snapshot child; `SHELL=/bin/sh` works around it
- Cloud sessions (`/ultraplan` handoff) die on first turn with generic "An error occurred while executing Claude Code" when repo-tracked settings enable sandbox with `failIfUnavailable: true`
- [Bug] Usage credits not updating after adding new credits from another machine
- Workflow tool: args reaches the script as a JSON string even when the caller passes a real JSON object
- setup-cowork étape 5 référence un skill setup-writing-style qui n'existe pas
- [Bug] Claude Code falsely triggers Fable 5 safeguard on legitimate QA/security testing documentation
- [Bug] Unexpected API charges and billing discrepancy investigation needed
- [Bug] Claude Code executes unapproved commands on user infrastructure without explicit confirmation
- claude.ai connector MCP session silently wedges after ~15 min idle — no drop detected, no reconnect; fresh process succeeds on the same call simultaneously
- [Bug] Team Plan 50% Weekly Limit Promotion Not Applied to Account
- [Bug] Excessive refusal behavior blocking legitimate code modifications
- [BUG] Windows: Desktop MSIX stuck Staged/Servicing — survives full OS repair install AND Developer Mode; installer fails 0x80073D28 (no CoworkVMService)
- Exact string "[Request interrupted by user for tool use]" returned on a FOREGROUND Task/subagent dispatch with no user interrupt (CLI, macOS, v2.1.212)
- [Feature Request] Support for Level 4 Autonomy across portals, servers, and devices
- Project-scope agent in .claude/agents/ intermittently fails to register at session start (same file, same version — nondeterministic)
- [BUG] --from-pr shows "No conversations found" for existing linked sessions: PR filter only applies to loadedpicker pages
- are [Bug] Anthropic API Error: Safety measures blocked cybersecurity content processing
- [Bug] Model Performance Regression: Opus 4.5-4.6 Quality Degradation and Fable Hallucinations in Agentic Tasks
- [FEATURE] Expose effort as a per-call parameter on the plain Agent tool (not just Workflow's agent())
- Anthropic: Go fuck yourself with your safeguard!!
- [Bug] /login command causes infinite redirect loop on authorization page
- [Bug] File modification timestamps causing continuous code changes during hardening operations
- Integrated browser: unattended JS dialogs (alert/confirm/prompt) queue up and freeze the desktop app
- [Feature Request] Support for direct audio input from WhatsApp integration
- [BUG] Claude Desktop doesn't expose claude.ai cloud connectors to the model despite being toggled on
- Make double-press Ctrl+C/Ctrl+D exit confirmation configurable
- Session limit still enforced after plan upgrade — upgrade doesn't lift the cap it was purchased to remove
- Session-limit interactive prompt blocks queued scheduled prompts (CronCreate//loop) in unattended sessions
- [BUG] Permission approval prompts never appear in the voice/text interface
- /insights: large-session facets record a 500-token error that never occurred, misfiling them as unclear_from_transcript
- Remote Control never connects: "Cannot read properties of undefined (reading 'session_url')" on desktop app
- [Feature Request] Add visual diff/comparison mode for iterative UI adjustments
- [BUG] Persistent/duplicated safety policy text being injected into unrelated Claude Code sessions, consuming tokens and burning through usage limits
- [Bug] Plugin hooks silently dropped when plugin has no version
- [Bug] Anthropic API Error: Model access denied for claude-fable-5
- Remote control: disconnect throws "Cannot read properties of undefined (reading 'session_url')" after code 4090 takeover
- [BUG]
- Desktop: /cd mid-session splits transcript across project dirs — "No conversation found with session ID"
- I need more information to generate a GitHub issue title. Could you please provide details about the bug or feature request you're experiencing with Claude Code? Include: - What were you trying to do? - What error or unexpected behavior occurred? - Any er
- [BUG] The worst experience I have with your app. Claude code is really very very bad in app. Never use app for code if you don't want to be mad
- I need more information to create a specific GitHub issue title. Could you please provide: 1. What specific bug or problem are you experiencing? 2. What error messages are you seeing (if any)? 3. What behavior is occurring that shouldn't be, or what's mis
- [BUG] Claude Desktop (MSIX) no longer launches after updating to 1.22209.0.0 — processes run but no window, on AD domain machine with OneDrive folder redirection
- Add confirmation before closing session with double Ctrl+C
- loop trying to login.
- [Bug] Fable 5 safeguards triggering false positives on literary criticism terms
- [Bug] False positive error when building UI
- Feature request: SubagentStart hook stdin should carry the Task tool description for subagent attribution
- [DOCS] Permission docs do not distinguish dir/** scope for allow versus deny and ask rules
- [DOCS] Tools reference is missing the EndConversation tool
- [DOCS] OpenTelemetry log-event reference omits new message correlation and tool provenance attributes
- [DOCS] OpenTelemetry documentation does not explain CLAUDE_CODE_OTEL_CONTENT_MAX_LENGTH
- [DOCS] subagentStatusLine documentation omits the new reasoning-effort payload field
- [DOCS] Ultrareview documentation omits the no-merge-base fallback
- [DOCS] Hooks reference omits the SessionStart “fork” source
- [DOCS] Hook if-condition documentation omits the changed scope of single-segment dir/** patterns
- Browser pane: tabId rejected as missing despite documented "default: the fronted tab" (read_page, javascript_tool)
- [Bug] Anthropic API Safeguard False Positive on Academic Content in Markdown Editorial Review
- [BUG] Sidebar shows "Running" for a session that has already completed
- Local stdio MCP server subprocesses leak indefinitely across session lifetime
- OpenTelemetry: bridged / remote-controlled sessions emit no local OTel metrics
- [FEATURE] Feature request: persist or let me replay the rotating spinner-tip hints
- [Bug] Claude hallucinates DMTCP as cybersecurity software instead of checkpoint-restart tool
- Headless -p mode hangs forever reusing a server-closed idle keep-alive connection after a long tool execution
- Agentic session (subagent-driven development + systematic-debugging) fixed a bug but didn't catch that its own alerting had been silently broken for 5 days
- [Bug] Anthropic API Error: False positive safety filter on PDF text extraction
- Idle TUI sessions burn 10–40% CPU when an animated element stays on screen
Resolved issues (176)
- (withdrawn)
- You've hit your session limit
- [Bug] Excessive token usage and incomplete incremental responses
- [Bug] MCP tool permission rules in allow list not suppressing approval prompts in auto mode
- [BUG] Claude Code crashes with 0xC0000005 on Windows 11 when using Bun-based versions (v2.1.112+)
- [Bug] Tool use rejected signal causes loss of AskUserQuestion answers
- [FEATURE] First-class steering queue UI: view, delete, and inject queued messages individually
- [VS Code Extension] "Unsupported content type: server_tool_use" shown repeatedly in chat transcript
- Desktop app and remote-control sessions never render hook `systemMessage` (PreToolUse/PostToolUse)
- Let the editor-tab title read from the session summary line for slash-command-launched sessions (VS Code)
- [FEATURE] Command execution prompts should explain what the command does and why
- [BUG] VS Code extension 2.1.211: renderer crash (SIGBUS, code 135) opening projects with large session-transcript dirs
- [BUG] Claude Desktop: can't load attachments or read pastes-as-attachments after forced update to 1.21459.0
- [BUG] Claude desktop didn't reset quota for each 5 hrs even do nothing around 2 days.
- [BUG] Desktop Connector Install/Uninstall button is disabled after upgrading Claude Desktop to v1.21459
- MCP connection uses stale cached settings (host/port) instead of current config, even after full app restart
- Claude Desktop repeatedly re-archives sessions in bursts, ignoring ccAutoArchiveOnPrClose: false
- [BUG] Claude Tag recurring trigger fails with session_config_rejected — dead environment id auto-assigned
- [BUG] Filesystem extension install crashes Extensions.getInstalledExtensionsWithState (TypeError: u._parse is not a function) — macOS
- [Bug] License validation QA falsely flagged due to security terminology
- [Bug] Security flag triggered incorrectly outside cybersecurity context
- [Bug] Harmless message incorrectly rejected/not accepted
- [Bug] Excessive unnecessary agent calls in agentic loop
- [Bug] Model routing error: Fable 5 incorrectly routed to Opus 4.8
- [Feature Request] TUI fullscreen mode not functional in VS Code
- [Bug Report: Unclear issue description - please provide details]
- I don't have enough context to generate an appropriate GitHub issue title from your message. Could you please provide: 1. What specific bug or feature is this about? 2. What error message or unexpected behavior occurred? 3. What were you trying to do when
- [Bug] Model unexpectedly downgraded from Claude 3.5 Sonnet to Claude 3 Opus
- claude-in-chrome
- [Bug] Anthropic API Error: Overly Restrictive Content Policy on Abstract Mathematical Simulations
- Feature request: Codex-style background computer control with concurrent window/session ownership
- Feature Request: Cross-device session continuity — connect to active Claude Code sessions from any Claude client
- [Feature Request] Allow custom naming for git worktrees instead of random generation
- [Bug] Unable to process empty bug report
- [BUG] Welcome banner shows subscription default model instead of settings.json-pinned model
- [BUG] Claude Code activity in organisation analytics are missing some users
- [Feature Request] Whitelist repository for standard cellpose/cupy 3D cell segmentation workflow
- [Feature Request] Allow programmatic/agent-initiated compaction (refile of #52002)
- [Feature Request] Allow Claude Code to assist with mold detection model development and validation
- [BUG] Output content filter blocks legitimate code-signing build scripts (false positive on Authenticode / Microsoft Store signing)
- [Bug] Anthropic API Error: Server rate limiting during request
- [Feature Request] Add model selection option to specify preferred model for code generation tasks
- [Bug] Biology Safety Filter triggering prematurely on health data analysis
- You've hit your session spend limit · raise it at claude.ai/settings/usage
- [Bug] Weekly usage limit reset occurring before scheduled reset time & new week ends in 5 days, not 7 days
- [BUG] pnpm lifecycle-script spawn rejected with EPERM under macOS sandbox; Node-direct spawn of same shell succeeds
- [Feature Request] Improve topic classification for creative/artistic requests
- [Bug] Missing or invalid bug report content
- [BUG] Black floating rectangle artifact follows cursor when Claude desktop app is open (macOS)
- False positive: implementing authentication for one's own app flagged as a security concern
- [FEATURE] Configurable diff background colors / support terminal theme colors
- [Bug] Fable 5 bio classifier blocks innocuous academic statistics terms (组学, spatial transcriptomics)
- $schema const locks to stale SchemaStore URL — local/updated schema not accepted
- [FEATURE] PreResponse/PostResponse hook for assistant text output (MessageDisplay is display-only)
- [Bug] Claude Code security review false positive on CORS origin validation code
- [BUG] plugin marketplace update/install fails for never-launched users — official marketplace auto-seeded only on interactive first launch, no autoheal
- [BUG] Native installer: in-memory buffering + no progress + 10-min hard timeout + restart-from-zero retries make updates impossible (and look like a deadlock) on slow routes to downloads.claude.ai
- [BUG] Fable 5 safety classifier auto-switches to Opus mid-session during authorized, in-scope security testing
- [BUG] Fable 5 repeatedly spawns verify subagents in `/effort ultracode` workflow mode and exhausts session limit
- [Bug]: Fable 5 flags benign greeting ("hi") due to security-related filenames in system-injected git context
- [BUG] Session usage metrics inconsistent between Claude Code app and web
- [Bug] Model Fallback: Fable 5 (1M) reverts to Opus 4.8 instead of Opus 4.8 (1M)
- Ultracode mode can silently consume an entire 5-hour usage window on subagent fleets
- I appreciate you reaching out, but I need to clarify something important: **This is not the right channel for access requests or account issues.** This tool is for reporting technical bugs in Claude Code. Your message appears to be a request for access o
- [Bug] Anthropic API Error: Overly broad content filtering on technical architecture questions
- [Bug] Anthropic API Error: Incorrect content safety flagging for gardening documentation
- [BUG] Model responded to a user message that does not exist in the session JSONL ("write a letter to the next version of you" injected mid-session)
- Workflow: write agents write status reports instead of file content; schema-read agents silently return null; resumed workflow overwrites manual fixes after long delay
- [Bug] Anthropic API Error: Overly Restrictive Content Policy Blocking Legitimate Username Collision Detection
- API 400: harness emits content block with type: "fallback" — session becomes permanently unrecoverable
- [Bug Report] Unable to process empty bug report
- [FEATURE]
- [BUG/ENHANCEMENT] Hooks added to settings.json mid-session aren't enforced until restart; no durable session-scoped hook state
- Feature request: keep Fable 5 in the model picker after June 23, billed via extra usage credits
- [BUG] Tool calls intermittently emitted as literal text (stray "count" + raw <invoke>) instead of executing — Opus 4.8, long Edit-heavy session
- [FEATURE] Allow voice dictation while AskUserQuestion option picker is shown
- [Bug] Legitimate open source development tasks incorrectly blocked by Cyber Safeguards (False Positive)
- Agent Teams: Spawned teammates don't inherit model configuration from team lead
- [BUG] FleetView PR chips never clear — keyed by branch name, collide across repos
- [Bug] Safety filter incorrectly flags population statistics as potential misuse
- [FEATURE]
- [Feature Request] Allow bio-inspired learning topics without biology classification blocking
- [Bug] Authentication failure with rotated 1Password token causes session interruption and model downgrade
- Agent viewer re-renders session welcome banner, mislabeling subagents' model (v2.1.214)
- MCP OAuth state mismatch on Windows — root cause likely matches closed #17733 (state/redirect_uri omitted from authorize request)
- MCP client does not re-fetch tools/list on notifications/tools/list_changed — new tools unreachable until full session restart
- [BUG] Cowork MCP client strips tool inputSchema properties typed as nullable unions (["string","null"]) — enums and nested schemas never reach the model
- MCP: tools/list fails ("tools fetch failed") for servers with recursive $ref in outputSchema
- MCP HTTP client fails to resolve $defs/$ref in tool schema ("can't resolve reference")
- plugin:engineering:github MCP fails to connect — dynamic OAuth client registration incompatible with GitHub Copilot MCP endpoint
- [BUG]
- skill-creator eval/optimizer leaks MCP child processes via headless 'claude -p' — exhausted memory, forced hard reboot
- Mid-turn channel-message injection: no way to declare an operator channel as trusted ("This is NOT from your user")
- [BUG] Desktop 1.21459.0 (macOS): Extensions page fails with getInstalledExtensionsWithState TypeError: u._parse is not a function whenever any extension is installed
- [Billing BUG] Max 20x paid but account provisioned as 5x
- Claude Desktop v1.21459.0: bootstrap-delivered config suppresses deployment-mode chooser (Claude.ai login no longer offered)
- Filesystem connector unavailable after update to v1.21459.0
- Extensions/Connectors screen crash after 1.21459.0 update: getInstalledExtensionsWithState — "u._parse is not a function" (Windows, org allowlist)
- [FEATURE]
- [BUG] Windows: Claude Code sessions and chat switching in Claude Desktop cause system-wide cursor stutter whenever the window is focused — MSIX blocks the GPU compositing workaround
- [BUG] Desktop (Windows): "Download and open in VS Code" silently fails for .py files — blocked by executable extension filter
- comiten und puschen
- [Bug] Claude Code enters infinite loop with repeated false positive "fix is in" declarations and excessive processing delays
- [Bug] Excessive token consumption on small projects with incomplete analysis
- I don't have enough information in this request to generate an appropriate GitHub issue title. You've provided only a Request ID, but I need an actual bug report or description of the problem to create a meaningful issue title. Please provide: - A descrip
- [Bug] CVP-approved account falling back on security runs
- [Bug] Safeguard incorrectly flags legitimate herbalist content
- [All models] Claude ignores safety-critical blocking questions and proceeds without explicit answers
- [Bug] Anthropic API Error: Server rate limiting requests
- [Bug] Anthropic API Error: Server Rate Limiting During Normal Usage
- [Bug] Security safeguard incorrectly flags user's own source code during audit
- [Bug] Security-guidance plugin bypasses subscription billing and uses expensive model without warning
- [BUG] Cowork: Linux bash sandbox never finishes booting on macOS — every command returns "Workspace still starting"
- [BUG] Voice dictation fails ("unstructured error frame", WS close 1008 Invalid authorization) when authenticated via CLAUDE_CODE_OAUTH_TOKEN — voice WebSocket opened without Authorization header
- Bash tool: injected grep() shell function kills the whole tool shell on no-match — compound commands silently truncate
- [Bug] Cursor position rendering error in iTerm2 after v2.1.170 upgrade
- [BUG] Claude Button Disappears When Files Close
- I'm Claude Code, Anthropic's CLI tool for coding tasks. I'm not able to help with bug bounty programs or security testing matters. If you've found a legitimate security vulnerability in Claude Code itself, please report it responsibly through Anthropic's
- [FEATURE] Configurable Spinner
- [Feature Request] Allow Claude Code use for peptide design and scientific research workflows
- [FEATURE] Subtractive exclude-list for spinner verbs (remove one verb, keep the curated defaults)
- [BUG] Repeated "image could not be processed" errors consumed ~60% of usage limit — 19 consecutive failures, each resending full 34MB context
- [BUG] Limits bug
- I don't have enough context to generate a GitHub issue title from your message. "Why fable flagged?" doesn't provide clear details about: - What the actual problem is - What error occurred - What you were trying to do - What the expected vs. actual behavi
- [Feature Request] Add usage telemetry to distinguish legitimate security hardening from malicious intent
- [Bug] Silent model fallback from Fable 5 to Opus 4.8 overrides explicit model pinning
- [Bug] Model downgrade blocks legitimate security auditing workflows
- [Feature Request] Improve edge case handling during code debugging
- [Bug] Claude Code: Fable5 model not responding to security audit requests
- [Bug] Unexpected model switch to Claude Opus 4.8 during Rust/Go development
- [BUG] Auto-scroll in VSCode extension is too aggressive — gets stuck pulling user back to bottom
- [CRITICAL BUG] Desktop auto-update wipes ~/.claude/projects/ — 1 month of sessions permanently deleted, no warning
- [FEATURE] Install location option for claude install #21019
- Permission rules are bypassed by semantically-equivalent command%2 0forms (`git -C`, `cd &&`, env prefixes)
- [BUG] ● API Error: 400 tools.13.model: 'claude-opus-4-8' cannot be used as an advisor when the request model is 'claude-fable-5'.
- [BUG] undefined is not an object (evaluating 'T.data.length')
- Claude Code (Opus): self-favoring asymmetric skepticism, unstable calibration, and a false-completion claim across one session — the session transcript is the evidence
- [BUG] Pro subscription on web but shows "Free" in desktop app on macOS Ventura 13.7.8
- I don't see a bug report in your message—only a request ID (`req_011Cbu81SJXEAGUokMxMTSir`). Please provide the actual bug report details, including: - What the issue is - Error messages (if any) - Steps to reproduce - Expected vs. actual behavior Once
- [BUG] Cowork Windows: CoworkVMService crash (SCM event 7034) with no recovery action; claude.exe hangs against the dead service for 3+ hours (v1.11847.5.0)
- [Bug] Model reverts to Opus unexpectedly during code review
- enabledPlugins in settings.json not persisting across restarts
- /effort without arguments should show interactive slider UI
- [MODEL] Bug， Non-existent content was inserted after multiple rounds of dialogue.
- Sonnet 4.6 1M and Opus 4.8 are fucking dumb
- [BUG] Cloudflare WAF blocks OAuth token refresh from headless Linux servers — permanent lockout after ~1 hour
- [Feature Request] Self-service toggle for individual extra usage opt-in
- Skills from additional directories (add-dir) not suggested in start agent dialogue
- [Bug] Anthropic API Error: Over-aggressive content filtering blocking legitimate code review requests
- [BUG] Account permanently bricked: every request fails with "invalid high surrogate in string" at fixed position — corrupted server-side account state
- New Session button is not working on Mac Air M4 Tahoe
- False positive: Usage Policy block on lattice gauge theory question Fable 5, v2.1.170) [BUG] Request ID: req_011Cbt6Zpk3Rup3RBX4CjmLq
- [BUG] [VSCode Extension] Plan content not visible in panel before accepting plan mode prompt
- Usage creeping up unexpectedly while idle; session reports unfamiliar model "Fable" (claude-fable-5[1m])
- Feature Request: Scope Claude Code to a subfolder within a multi-folder workspace
- [Bug] claude-opus-4-8 corrupts tool-call boundary token, emitting raw XML instead of tool_use blocks
- [BUG] login issue through terminal and vs code on a Mac
- [BUG] JetBrains Plugin: Input area shows only 2 lines and text unreadable in Remote Development (Gateway)
- Improve sandbox support for erlang / elixir
- [BUG] Desktop "Update" button fails for plugins installed at managed scope - CLI update invoked at user scope
- Feature request: blocking acknowledgment hook for /clear (and other slash commands)
- [DOCS] VS Code docs should clarify `/usage` opens the Account & Usage dialog
- Skill tool: tab-completed short skill name fails with 'Unknown skill' when invoked mid-message
- Bug: Claude Desktop fails to find downloaded VM files and workspace initialization fails
- Security audit on my website githubb repo flagged by mythos
- [BUG] When using Foundry, Test connection is trying to connect to /models/chat/completions instead of anthropic/v1/messages
- /rename command does not persist in /resume list
- Functional Optimization
- [FEATURE] Support virtual filesystems (non-file:// URI schemes) in VSCode extension
- [Bug] Overly aggressive safety guardrails blocking legitimate code security audits
- New Google Chrome Beta Extension causing issues with website language content
- [BUG] Max 20x subscription silently showing as Pro — backend account sync issue
- Cowork: keep a chosen live artifact pinned in the viewer during, across sessions and relaunch (and don't let file previews evict it)
- Remote session viewer shows stale 'running' tool tasks from dead sessions; dismissal does not persist
- [Bug] Goal function causes infinite loop when waiting for task completion
- [BUG] HCS operation failed: failed to start VM: HcsWaitForOperationResult failed with HRESULT