Security Good First Issues

Browse open source issues grouped under Security. Where the subject goes by older names in the issue metadata, those are matched too.

Security

Slice signals

Average annotated difficulty: 4.1/5 across 4,479 issues.
7% of 4,479 time estimates are one day or less.

Open All

[Bug] Interactive auto mode bypasses permissions.ask rules for PreToolUse decisions Open

area:hooks area:permissions bug has repro platform:linux

anthropics/claude-code#87639 · 1 comment · 1 reaction · 0 assignees · Indexed Sep 14, 2026
[Bug] Fable 5 safeguard `[reasoning_extraction]` false-positives on a one-word greeting ("Hi") Open

area:model bug platform:macos

anthropics/claude-code#87640 · 13 comments · 15 reactions · 0 assignees · Indexed Sep 14, 2026
Model fabricated a fake `user` turn and `system` block mid-response, then denied authorship; also refused an explicitly authorized deployment and the bug report about it Open

area:model bug platform:vscode platform:windows stale

anthropics/claude-code#87670 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Bash sandbox fails to initialize on Ubuntu 24.04+ - conflict with the distro's default bwrap-userns-restrict AppArmor profile (not a Claude Code regression) Open

area:bash area:sandbox bug has repro platform:linux stale

anthropics/claude-code#87680 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] mcp-needs-auth-cache.json is shared across CLAUDE_SECURESTORAGE_CONFIG_DIR profiles — one account's auth failure blocks the other's MCP servers Open

stale

anthropics/claude-code#87702 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Desktop 1.32352.x: first-time OAuth MCP connect never opens the browser — version-negotiation probe wraps a client-side auth error, which is then misclassified as transport Open

area:desktop area:mcp

anthropics/claude-code#87713 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Fable 5 safeguard false-positives derail long agent sessions Open

area:model bug stale

anthropics/claude-code#87718 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
auth login: code prompt does not echo, never says to press Enter, and empty submission returns the same 400 as a bad code Open

stale

anthropics/claude-code#87731 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Anthropic API Error: Excessive content policy flagging blocking legitimate requests Open

area:model bug platform:windows stale

anthropics/claude-code#87736 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Desktop Browser pane has no way to use existing browser login state — no profile/cookie import, and the Chrome extension doesn't fill the gap Open

area:desktop enhancement

anthropics/claude-code#87764 · 2 comments · 1 reaction · 0 assignees · Indexed Sep 14, 2026
[Bug] CVP approval not recognized across organizations in Claude Code Open

area:auth area:security platform:linux question stale

anthropics/claude-code#87766 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] ask permission rule for Bash(git push *) not triggering confirmation prompt Open

area:permissions bug platform:macos stale

anthropics/claude-code#87773 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Permission/privacy messages say THAT files are accessed, but not WHICH — collapsed UI hides the target path Open

area:tools area:tui duplicate platform:macos stale

anthropics/claude-code#87775 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Dual-use safeguard false positives on legitimate defensive security review; mid-session auto-model-switch breaks continuity Open

area:model area:security duplicate platform:macos

anthropics/claude-code#87802 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Anthropic API Error: Account validation state mismatch for forensics access Open

area:model bug platform:windows stale

anthropics/claude-code#87803 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Jammed background tasks + Remote Control reconnect loops silently consume Max usage window after forced token rotation Open

area:agent-view area:auth area:cost bug platform:macos stale

anthropics/claude-code#87805 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Permission classifier blocks the actions that would grant permission — no escape hatch in non-interactive sessions Open

area:desktop area:permissions bug platform:windows

anthropics/claude-code#87809 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Overly broad safeguard triggers false positive on legitimate internal IT administration and network diagnostics Open

area:model bug platform:macos stale

anthropics/claude-code#87816 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Bug report: assistant fabricated a user turn and system prompts inside its own response, then executed them Open

api:anthropic area:model area:security bug platform:linux platform:vscode

anthropics/claude-code#87823 · 4 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Persistent memory rules are decorative — Claude ignores them before destructive actions, repeatedly, across sessions Open

area:model bug memory platform:macos

anthropics/claude-code#87825 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Starting a session in Claude Desktop revokes filesystem access from already-running CLI sessions (macOS TCC identity collision) Open

area:desktop area:security bug has repro platform:macos

anthropics/claude-code#87833 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
claude mcp get/list print configured MCP secrets (headers, env vars) in cleartext with no masking Open

area:cli area:mcp area:security enhancement stale

anthropics/claude-code#87838 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG]Scheduled task stalls indefinitely on Write permission prompt despite matching allow rule (settings.json correct) Open

area:permissions area:skills bug platform:macos stale

anthropics/claude-code#87843 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Feature Request] Add legitimate coding pattern recognition to reduce false positive security flags Open

stale

anthropics/claude-code#87846 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Auto-mode permission classifier denies allowed operations silently with incorrect remediation guidance Open

area:permissions bug has repro platform:linux reproduced

anthropics/claude-code#87850 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Desktop app browser has no system password manager integration (Apple Passwords autofill unavailable on macOS) Open

area:desktop

anthropics/claude-code#87853 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Desktop app browser does not retain sessions despite session persistence set to shared Open

area:desktop bug platform:macos stale

anthropics/claude-code#87854 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Prompt Injection: User utterances can trigger unintended code execution Open

stale

anthropics/claude-code#87864 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Model emitted a standalone assistant turn impersonating a user instruction (asked to bypass review and push to prod) Open

area:model area:security bug has repro platform:macos stale

anthropics/claude-code#87871 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[FEATURE] Auto mode classifier should evaluate declared intent, not just the action Open

area:permissions

anthropics/claude-code#87881 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Feature Request] Add option to configure security checks behavior in Claude Code CLI Open

stale

anthropics/claude-code#87886 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Anthropic API Error: Cyber Safeguards Blocking Legitimate Requests Open

api:anthropic area:model bug platform:linux stale

anthropics/claude-code#87903 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Anthropic API Error: Opus 5 Safeguards Flagged Legitimate Request Open

area:model bug platform:vscode platform:windows stale

anthropics/claude-code#87907 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Feature Request] Add CTF/security testing mode for controlled vulnerability exploitation Open

area:model area:security enhancement stale

anthropics/claude-code#87908 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[FEATURE] `claude auth login --email` should force the OAuth account chooser (prompt=select_account) instead of reusing the browser's claude.ai session Open

area:auth area:cli enhancement

anthropics/claude-code#87922 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Sandboxed Bash wedges permanently when the working directory is on a read-only filesystem (no cwd recovery) Open

area:bash area:sandbox bug has repro platform:linux reproduced stale

anthropics/claude-code#87935 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] cyber safeguard model-fallback is now session-scoped (2.1.229) and false-positives on legitimate security work — reproduced live on Opus 5 Open

area:model area:security bug has repro

anthropics/claude-code#87938 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Browser-pane dev-server launcher fails with EPERM (uv_cwd) -- sandboxed subprocess can't load an ad-hoc-signed native binary (Vite 8/Rolldown) Open

area:sandbox bug has repro platform:macos stale

anthropics/claude-code#87939 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Personal Pro subscription blocked in Claude Code — "organization has disabled subscription access" (auto-generated personal org) Open

area:auth bug duplicate platform:windows

anthropics/claude-code#87942 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Anthropic API Error: Overly Aggressive Content Filtering on Legitimate Cybersecurity Development Work Open

area:model duplicate external platform:macos stale

anthropics/claude-code#87949 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Cowork forces a full account logout (auth_kind=session_stale_relogin) roughly every 24-48h on Linux desktop Open

area:auth area:cowork area:desktop bug has repro platform:linux

anthropics/claude-code#87950 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 14, 2026
Worktree-isolation Bash guard refuses every compound command, even with no git usage at all Open

area:bash area:sandbox enhancement platform:macos

anthropics/claude-code#87959 · 5 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Model fabricates a "prompt injection" incident — quotes a fake attack string that never appears in any tool_result Open

api:bedrock area:model bug stale

anthropics/claude-code#87965 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Pre-commit hook interaction leads Claude Code to rm -rf files from a concurrent, unrelated session Open

area:hooks bug data-loss platform:macos

anthropics/claude-code#87981 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Background subagents (e.g. via /code-review) silently auto-deny gated Bash commands instead of prompting Open

area:agents area:permissions bug has repro platform:windows stale

anthropics/claude-code#87994 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[Bug] Cybersecurity classifier blocking defensive malware analysis tooling mid-conversation Open

area:model area:security bug platform:macos stale

anthropics/claude-code#88014 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
[BUG] Claude Code fails to authenticate on MacOS in certain environments. Open

area:auth area:security bug has repro platform:macos

anthropics/claude-code#88024 · 1 comment · 1 reaction · 0 assignees · Indexed Sep 14, 2026
Write tool does not enforce the working-directory sandbox that Bash enforces Open

area:permissions area:sandbox area:security bug has repro platform:macos stale

anthropics/claude-code#88038 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
Worktree command-safety analyzer misreads a non-leading `complete`/`compgen`/`compopt` token as an invoked builtin Open

area:sandbox bug has repro platform:windows stale

anthropics/claude-code#88047 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026
`claude remote-control` server exits on 401 after exactly 24h — does not refresh its OAuth access token, killing every attached session Open

area:auth bug has repro platform:macos

anthropics/claude-code#88054 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 14, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

Security Good First Issues

Slice signals

Showing the newest 100

Get new issues in your inbox