Agents issues on macOS — page 2
Issues the maintainers labelled both platform:macos and area:agents.
880 issues · 266 open · 614 resolved (70%) · first seen Jul 25, 2025
Is this getting better or worse?
This class of problem is holding steady. 415 new reports in the last 90 days vs 429 in the 90 before — -3%. The open backlog peaked at 266 in 2026-08 and sits at 266 today.
Derived by counting each matching issue from its creation date until it closed. First-party
computation over the anthropics/claude-code tracker — the trend is not published
anywhere upstream.
Which builds report this
632 of these reports name the Claude Code build they were running, spanning 146 releases. Heaviest reporters:
- Claude Code v2.1.119
- Claude Code v2.1.233
- Claude Code v2.1.201
- Claude Code v2.1.198
- Claude Code v2.1.197
- Claude Code v2.1.126
- Claude Code v2.1.207
- Claude Code v2.1.139
Report volume, not a defect count: an issue can mention a build without that build having caused the failure. Versions come from the version string in the bug-report template, matched against real published tags.
Do these get fixed?
Median time to close is 32 days across 614 closures. Of the 614 closures with a recorded reason, 24% were closed as completed and 466 as not-planned or duplicate. The most recent completed fix landed in 2026-08.
Workarounds reported by the community
267 of these issues post a workaround someone says worked. The highest-engagement ones:
- Built-in Plan agent ignores parent settings.json permissions and repeatedly prompts for pre-approved tools
- …agent permissions (e.g., ) 3. Option C : Add a setting like Workarounds Attempted None available. Cannot modify built-in agent configurations. Impact - Interrupts workflow with repeated permission prompts - Makes Plan mo…
Found in the description of #10906 · still open - [BUG] Sub-agents claim successful file creation but files don't persist to filesystem
- …nsparency issues - 3408 - Task tool crash scenarios Current Workarounds - Use direct tool calls instead of Task tool for file operations - Always verify file creation externally after sub-agent operations - Avoid sub-age…
Found in the description of #4462 · still open - [Bug] Regression: malware reminder on every Read still causes subagent refusals in v2.1.111 (fix from #47027 / v2.1.92 did not hold)
- …ain to the subagent that it's misreading" is not a reliable workaround — it can increase suspicion, mirroring the OP's subagent-2/3 reasoning chain. Per-agent inconsistency, same repo, same reminder, same session: two ot…
Found in the thread of #49363 · resolved - [CRITICAL] Subagent spawning and subagent pattern bugs trigger infinite recursion, infinite token usage, grossly inefficient token usage, and lost accumulated subagent work.
- …The agent ignores this and instead spawns a child agent to work around the denial. That child hits the same wall and spawns another. This repeats indefinitely. The permission denial is functioning as a recursive agent l…
Found in the description of #68619 · still open - [Bug] Plan agent uses main branch path instead of git worktree directory
- …an agent should use the current working directory Temporary Workaround In the meantime, when the Plan agent gives you a plan, you may need to: - Review the plan carefully for any incorrect path assumptions - Manually cor…
Found in the description of #11239 · still open
Excerpts are quoted from the linked issue so you can judge relevance before opening it. Threads that only ask for a workaround, or report that there is none, are excluded. These are community suggestions, not official guidance from Anthropic — read the full thread before applying one.
Matching issues 101–200
Ranked by community engagement (reactions weighted over comments). Listing the top 300 of 880 matches; the statistics above cover all 880.
- [BUG] Opus 4.8 fabricates user turns across 3 related sessions — one executed a real unauthorized commit+push on a fabricated confirmation (JSONL-verified)
- Subagent cache miss on first SendMessage resume (cache_read=0)
- /agents command shows 'No agents found' despite valid agent files in .claude/agents/
- TeamDelete fails after agents approve shutdown - 'active members' not cleared
- Custom agents are undiscoverable via ToolSearch when Agent tool is deferred
- [BUG] TUI model badge for background subagents shows the session model, not the resolved override
- [BUG] Contradictory 'session limit resets at X' times across one session; background subagents die terminally and reset projections silently roll forward
- [Feature Request] Add per-subagent observability (model, effort) to /agents Running tab and agent JSON output
- [BUG] Claude Max weekly usage depleted unexpectedly fast for Claude Code subagent-heavy Gmail MCP sessions
- Background sessions (claude agents) drop work-phase OTEL logs (user_prompt/api_request) on shutdown
- Agent Teams: shutdown_request approval doesn't terminate teammate; reply delivery occasionally drops content
- [BUG] Cowork mode: sub-agent spawning completely broken — "Prompt is too long" at 0 tokens for all models
- [bug] Experimental agent team mode: "q.toolUseContext.getAppState is not a function" when sub-agent triggers permission prompt
- Subagent model resolution strips [1m] context window suffix
- [BUG] /batch command "API Error: Request rejected (429) · Rate limited" - Max 20x
- Remote Trigger /run returns HTTP 500 — trigger auto-disables after failed cron execution
- PreToolUse hook `updatedInput` silently ignored for Agent tool
- [BUG] Unsupported (?) model in agent config silently fails instead of erroring
- /code-review regression since v2.1.235: background forked-skill execution — parent stalls on last angle, and ReportFindings UI is lost
- [BUG] /agents not working in desktop app
- [Bug] Agent team tools not injected after auto-update to v2.1.178
- [BUG] Multiple background agents completing simultaneously overflow main context — API returns empty response
- Dispatch feature ignores CLAUDE_CODE_DISABLE_1M_CONTEXT — "Usage credits required for 1M context" blocks background agents
- Background tasks silently relaunched as duplicates + /tasks lacks elapsed time to triage
- Agent tool: spawned sub-agents go idle immediately without executing prompt
- [BUG] Subagent dispatches can silently fail when context-budget forcing function is injected — no error surfaced to orchestrator
- Background sessions block `Edit`/`Write` outside a worktree with no opt-out — gate is unconditional once `CLAUDE_JOB_DIR` is set
- [BUG]Advisor tool injects same-tier opus calls into every opus subagent, doubling spend invisibly
- [MODEL] Opus 4.7 regression: agents ignore CLAUDE.md directives, spawn in wrong worktrees, write code against stale assumptions
- [Feature Request] Add toggle keybinding to show running background agents with live output
- [BUG] Spawned subagent identifies as team-lead and exposes parent's conversation history when DM'd directly
- Subagent SendMessage delivery failures: Ink crash and silent tool-call drop in tmux teammate panes
- [BUG] Agent-Teams sub-agents with model: opus cannot call SendMessage / TaskCreate / TaskUpdate etc. despite frontmatter declaration (tmux backend)
- Subagent `skills:` frontmatter preload does not inject skill content when the skill has `disable-model-invocation: true`
- [Bug] Agent teams crash when team members request permission: getAppState is not a function
- [BUG] Subagents falsely report task completion without verifying file changes were written
- Worktree isolation breaks bare-repo worktree layouts by setting extensions.worktreeConfig
- Background tasks persist after /clear and cannot be stopped via programmatic Ctrl+C/Escape
- [BUG] subagents do not respect AWS_REGION/ANTHROPIC_SMALL_FAST_MODEL_AWS_REGION
- Parallel Agent tool uses count shows 0 when running 5+ agents concurrently
- [Bug] Excessive token consumption with multi-agent workflows using Sonnet models
- Background agents don't inherit Write/Edit tool allow rules from project settings
- [Bug] Parallel Agent calls can lose tool_results
- [BUG] Agent Teams: Teammate cannot exit properly in v2.1.76 (2.1.74 works, 2.1.75 broken)
- TeamCreate spawns teammates that silently exit due to incorrect command generation
- Subagents don't inherit session-level permissions
- [Bug] Global agents in ~/.claude/agents/ not recognized, only project-local agents work
- [Bug] Explore agent infinite loop or hang condition
- [BUG] Main agent not being able to read messages from sub agents (team agents)
- [BUG] SendMessage-resumed background subagent never appears in the background tasks UI
- [Bug] Uncontrolled recursive agent spawning causes excessive API usage and costs
- Sub-agent API 401 (auth expired) is reported to parent agent as "Request interrupted by user for tool use"
- [Bug] Subagents not awaiting nested subagent results, causing duplicate work and race conditions
- [Feature Request] Allow custom naming for git worktrees instead of random generation
- [BUG] Subagents return only "Done" with no output when dispatched for search/research tasks (Opus 4.8)
- [BUG] Workflow harness retries indefinitely on HTTP 429 rate_limit — 97 agents, 2M tokens burned in 34 seconds ()
- [BUG] Claude Opus 4.8 makes changes outside the current worktree for the current session
- Background `claude` subagent_type auto-wraps in git worktree; `general-purpose` does not; doc says neither should
- Background sub-agents intermittently return context-decoupled hallucinations (ignore prompt entirely, reported as 'completed')
- [BUG] agent thinking work lost when hitting limit
- Mouse-tracking escape sequences leak into prompt input from new agent view
- Agent tool: isolation: "worktree" creates worktree from default branch instead of parent HEAD
- [Bug] Unexpected high token consumption with multi-agent Opus orchestrator spawning Sonnet agents
- Crash: toolUseContext.getAppState is not a function when subagent triggers a Bash permission prompt
- Team agent progress visibility + keyboard shortcuts for Tasks/Plans tabs
- Subagents - their use and control and bug reporting via your chat bot
- Subagents should inherit settings.local.json permissions
- Sub-agent context isolation: parent MCP server registrations leak into sub-agent tool output
- Allow Claude to stop background agents.
- initialPrompt not auto-submitted when using --agent flag
- [BUG] Agent teams silently fall back to isolated subagents — runtime inconsistently chooses team path
- [BUG] Session crashes when worktree directory is deleted while Claude is operating inside it
- [BUG] Background task output files grow without limit, can fill disk after session crash
- [BUG] Claude code lose track of running subagents
- [BUG] Claude Code stopped respecting model in agent altogether
- Task output file grows unbounded — single file hit 107GB in 3 minutes
- [BUG] Explore subagent inherits parent model (opus) instead of using Haiku as documented
- Fable 5 safeguards false-positive [reasoning_extraction] terminates code-review subagents (mutation-testing vocabulary)
- Escape in the main input kills running background agents
- Feature request: a supported way to disable the agent_summary spinner (22.7% of requests, ~18% of weighted tokens in a subagent-heavy session)
- [BUG] Cross-session messages silently dropped since 2.1.227 — sender gets success, target session never receives them
- [Bug] Subagents billed after monthly spend limit exceeded; false clean review result on agent failures
- Agent tool subagents can access file paths outside approved permission boundaries
- Background-agent completion notifications silently dropped on 2.1.179 (sequential dispatch; agents complete cleanly, no task-notification fires)
- Subagent returns prompt-injection-shaped output with 0 tool uses (general-purpose, model: opus), localized to one long-running session
- [Bug] Workflow resume re-runs completed fetch agents instead of using journal cache
- PostToolUse hook does not fire for Agent tool completions
- [BUG] Ctrl+C stops interrupting in a long, heavy background-agent session (macOS TUI, v2.1.161); concurrent fresh sessions interrupt fine
- [BUG] Workflow subagents balloon to ~2GB each and aren't cleaned up — regression 2.1.156 (OK) → 2.1.160
- Feature Request: Configurable concurrent subagent limit in workflows
- [BUG] v2.1.154 regression: custom agents via `/agents` fail with "messages[1].role: unknown variant `system`"
- --bare mode should respect --tools flag to allow adding tools back (e.g. Agent, Skill)
- [BUG] agent view disabled by gatewayAuth gate on Max+Stripe accounts
- Runaway Agent task burns tokens after completion; agentId ≠ task_id makes it unkillable from parent session
- AskUserQuestion: Agentic processing breaks when sub-agent is asked to use AUQ by initial user-prompt
- [Bug] Auto-mode yields to user after Agent tool returns instead of continuing pipeline
- [Bug] Agent Teams launch crashes with Ink renderer: Box nested inside Text component
- [BUG] After agent used advisor the conversation got broken and keeps returning API Error: 400
- [Bug] Session hangs indefinitely on researcher tasks
- [BUG] Agent Teams: Lead agent requires manual stdin input to process teammate notifications — breaks unattended orchestration