Digest / February 18, 2026
Codex Issues Digest — February 18, 2026
92 new issues filed ·
27 resolved in openai/codex (UTC day).
New issues (92)
- False positive cyber-risk rerouting (routed to GPT-5.2)
- Codex App: New window isn't focused on launch
- feat(explorer-agents): built-in `input_context` tool — for iterative context pack building
- False positive cyber-safety rerouting (GPT-5.3-Codex → GPT-5.2 fallback)
- Flagged for high-risk cyber activity when not even using gpt-5.3-codex
- Typo in Orchestrator.md
- Cybersecurity Block for Benign Use
- /resume stopped working, codex freeze on it
- Flagged accounts - larger concerns
- Native Diff in VS Code file editor
- Codex App: up arrow command history isn't specific to current session
- False-positive “high-risk cyber activity” flag blocks gpt-5.3-codex in Codex CLI
- execution error: Io(Error { kind: InvalidInput, message: "batch file arguments are invalid" })
- App permissions change to 'custom' and causes read-only mode
- Getting routed to 5.2 after ID check.
- App auto scroll is buggy / broken
- Successfully completed trusted access verification; still blocked from gpt-5.3-codex
- False Positive - Flagged For Cyber Security Risk
- Codex repeatedly prompts for xcodebuild approval due to inconsistent shell wrapping (/bin/zsh -lc vs direct execution)
- Your account was flagged for potentially high-risk cyber activity and this request was routed to gpt-5.2 as a fallback. To regain access to gpt- 5.3-codex, apply for trusted access:
- Tabbed interface for parallel chat sessions in Codex extension
- Model metadata for `gpt-5-codex` not found. Defaulting to fallback metadata; this can degrade performance and cause issues
- Nix package broken since 0.89.0 version
- Upgrade to v0.103.0 causes unrelenting false positive cyber error message
- Account was flagged incorrectly
- Update loop
- Repeated Xcode permission prompt on every Codex app focus when using Apple Xcode MCP server
- Your account was flagged for potentially high-risk cyber activity
- [CODEX CLI] Rerouted to 5.2 from 5.3 even after succesfull cyber verification on PRO PLAN
- Voice input adjusts macOS mic input gain and does not restore it (mic level ends up near zero)
- Wrongfully flagged for potentially high-risk cyber activity
- Codex exec/CLI "Invalid Responses API request" error with OpenRouter
- Dynamically loading nested AGENTS.md
- Session does not persist across dev container rebuilds
- Login fails across multiple dev containers due to incremental port forwarding
- {"detail": "Bad Request"}
- IìMy account has been mistakenly flagged as high-risk cyber activity
- Deprecation warning links to removed docs anchor docs/config.md#feature-flags
- A false positive cyber-safety flag.
- Request Re-Routed for Cyber Risk False Positive
- Request Re-Routed for Cyber Risk False Positive
- `project_root_markers` is not honored for AGENTS.md discovery
- Make ENTER insert newline in Codex TUI input (send on Ctrl+Enter instead)
- False positive: Codex cyber-safety flag during normal UI/dev work
- cybersecurity flagging for benign gpt-5.3-codex usage
- GPT-5.3 requests automatically routed to GPT-5.2 under “Trusted Access for Cyber” (Pro subscription)
- Enter on terminal sends the prompt on the chat
- False positive: account flagged for cyber activity and routed to gpt-5.2 fallback
- Choose gpt5.2 xhigh maually but it sometimes change to medium during the execution
- False positive cyber-safety routing. gpt-5.3-codex usage
- False positive cyber-risk rerouting (re-routed to GPT-5.2)
- "The model `gpt-5.3-codex` does not exist or you do not have access to it."
- False positive cyber-risk
- False positive: routed to GPT-5.2 due to “cyber-abuse risk” during normal dev work
- High Risk Cyber Activity Bug
- False positive: account flagged for cyber activity and routed to gpt-5.2
- Your account was flagged for potentially high-risk cyber activity and this request was routed to gpt-5.2 as a fallback
- Your request was routed to GPT-5.2. - Cannot use GPT 5.3 Codex anymore
- SOLUTION - HIGH RISK CYBER SECURITY -> LOG OUT / LOG IN AFTER VERIFYING PERSONA
- Shell snapshot validation fails on macOS Bash (unexpected token '(') – v0.103.0
- Upgrade/add credits buttons don't work
- Thread ID 019c7161-0657-7bd1-b940-f315d6b2e03b inappropriately flags high-risk cyber activity for mundane style work
- Flagged for high-risk cyber activity when updating Codex config
- False‑Positive “High‑Risk Cyber Activity” Flag
- Locked out of using gpt-5.3-codex suddenly while doing nothing malicious, cyber website not working showing ineligible!
- 019c51ab-5625-7f03-994f-6aad706ceb7f False positive: Your request was re-routed to reduce cyber-abuse risk
- When using Codex, I encountered a problem where I needed to verify my identity via Trusted Access for Cyber. I uploaded my ID card, but I couldn't get through.
- Codex IDE keeps getting stuck on "Thinking"
- Thread title is auto-generated with malformed/error-like text (e.g. _invalid correction, Oops invalid JSON? I)
- False positive — account flagged for “potentially high-risk cyber activity” and rerouted to GPT-5.2
- macS app lost "Archived" threads permanently when trying to "Unarchive"
- account flagged for “potentially high-risk cyber activity” and rerouted to GPT-5.2
- Being downgraded to 5.2 | 019c6d28-add7-7b81-a15e-4fe448661006
- VS Code Codex extension repeatedly requires sign-in with ChatGPT auth, but works with API key auth
- Account being flagged for “potentially high-risk cyber activity”
- flagged for high risk cyber
- Bug: incorrectly flagged for cyber while building app; routed to GPT-5.2 Codex
- [Question] What did you guys do to make worktree creation much faster on codex desktop app
- Codex TUI treats scoped npm package strings as @ file mentions and blocks normal Enter submit
- Defaulting to original name in session renaming
- Graphik glitch
- Your account was flagged for potentially high-risk cyber activity and this request was routed to gpt-5.2 as a fallback.
- False Positive: High Risk Cyber Activity
- Multi-worker agent change question problem
- Impossible to turn off chatgpt.com and openai.com connections
- [BUG] collab deprecated
- Governance hooks: configurable policies, threat detection, and audit trails
- Randomly flagged for cyber security coding when doing research on HW products
- Expose skill origin and allow deterministic selection when skill names collide
- Codex Desktop App intermittently stuck on Thinking with anytls reuse=true (IpcRouter EPIPE)
- Add /mcp interactive TUI toggle (arrow + space) like /skills option 2
- Add “Clear all attachments” (bulk remove) in the composer
Resolved issues (27)
- Codex routing 5.3 codex to 5.2
- Wont start after update to 0.102.0
- Typo in Orchestrator.md
- Native Diff in VS Code file editor
- App permissions change to 'custom' and causes read-only mode
- chatgpt.com/cyber verified yesterday night, still being rerouted silently to 5.2
- Successfully completed trusted access verification; still blocked from gpt-5.3-codex
- GPT-5.3-Codex being routed to GPT-5.2
- I’m seeing unexpected server-side model routing in Codex CLI.
- False positive: gpt-5.3-codex rerouted to gpt-5.2
- Verified Pro subscriber silently routed to GPT-5.2 instead of requested GPT-5.3-codex
- Model routing mismatch: requesting gpt-5.3-codex returns gpt-5.2-2025-12-11 (verified via SSE response.created)
- GPT-5.3-Codex being routed to GPT-5.2
- Your account was flagged for potentially high-risk cyber activity and this request was routed to gpt-5.2 as a fallback. To regain access to gpt- 5.3-codex, apply for trusted access:
- Bug:Vscode Keyboard Shortcut "chatgpt.sidebarView.focus" work badly on the newest version of Codex plugin
- False positive cyber-safety rerouting (GPT-5.3-Codex → GPT-5.2 fallback)
- /resume stopped working, codex freeze on it
- Wrongfully flagged for potentially high-risk cyber activity
- Upgrade to v0.103.0 causes unrelenting false positive cyber error message
- Codex exec/CLI "Invalid Responses API request" error with OpenRouter
- Critical: Unexpected deletion/rename activity in personal folder (D:\Content) during Codex session, suggesting fail-closed path guardrails
- Choose gpt5.2 xhigh maually but it sometimes change to medium during the execution
- "The model `gpt-5.3-codex` does not exist or you do not have access to it."
- Update loop
- Being downgraded to 5.2 | 019c6d28-add7-7b81-a15e-4fe448661006
- Request Re-Routed for Cyber Risk False Positive
- npm install fails due to ENOTFOUND errors