Resume unloaded Codex threads with `codex queue --resume` and select GPT-6.1 Terra
Codex CLI rust-v0.166.0 adds GPT-6.1 Terra to bundled, Vertex AI, and Amazon Bedrock catalogs,codex queue --resume for unloaded threads, codex doctor --network-only WebSocket diagnostics, keyboard transcript selection with Find, and selectable safer TUI follow-ups after safety blocks.Why this loop
This release makes long-running Codex CLI work cheaper to recover and easier to steer. GPT-6.1 Terra is now in the bundled, Vertex AI, and Amazon Bedrock catalogs, so teams can pick it without a custom catalog. codex queue --resume resumes an unloaded thread after a queued message instead of opening a duplicate conversation. Resume and fork now keep acknowledged approval/permission settings and reasoning-token accounting, which avoids re-approving tools and premature compaction. Keyboard transcript selection (word, line, page motions plus Find and configurable shortcuts) plus consistent Markdown in async questions, choices, answers, and replies make review in the TUI practical. When a safety block hits, the TUI now explains it and offers selectable safer follow-ups. Run codex doctor --network-only first so configuration, connectivity, and WebSocket issues are isolated before queueing work; TUI reconnection recovery also handles offline /quit and /exit.
Proposed actions
- Upgrade Codex CLI to rust-v0.166.0, then set the session model to GPT-6.1 Terra from the bundled, Vertex AI, or Amazon Bedrock catalog.
- codex doctor --network-only
- After queueing a message onto an unloaded thread, run: codex queue --resume
- In the Codex TUI, enter keyboard transcript selection; move by word, line, and page; use integrated Find; set shortcuts if defaults conflict. Composer drafts stay intact while selecting.
- When a safety block appears, read the TUI explanation and pick a selectable safer follow-up prompt. Resume or fork the same thread—acknowledged approvals, permissions, and reasoning-token accounting are preserved.
Agent prompt
Paste into your agent or query via MCP (get_agent_prompt) — free, no extra AI costCodex / agent CLI task
DevAgentRadar → OpenAI Codex
Task: evaluate and optionally implement follow-ups from this coding-assistant release.
Context
Assistant: OpenAI Codex CLI
Proposal: Resume unloaded Codex threads with codex queue --resume and select GPT-6.1 Terra
Summary: Codex CLI rust-v0.166.0 adds GPT-6.1 Terra to bundled, Vertex AI, and Amazon Bedrock catalogs, codex queue --resume for unloaded threads, codex doctor --network-only WebSocket diagnostics, keyboard transcript selection with Find, and selectable safer TUI follow-ups after safety blocks.
Primary source: https://github.com/openai/codex/releases/tag/rust-v0.166.0
Why it matters
This release makes long-running Codex CLI work cheaper to recover and easier to steer. GPT-6.1 Terra is now in the bundled, Vertex AI, and Amazon Bedrock catalogs, so teams can pick it without a custom catalog. codex queue --resume resumes an unloaded thread after a queued message instead of opening a duplicate conversation. Resume and fork now keep acknowledged approval/permission settings and reasoning-token accounting, which avoids re-approving tools and premature compaction. Keyboard transcript selection (word, line, page motions plus Find and configurable shortcuts) plus consistent Markdown in async questions, choices, answers, and replies make review in the TUI practical. When a safety block hits, the TUI now explains it and offers selectable safer follow-ups. Run codex doctor --network-only first so configuration, connectivity, and WebSocket issues are isolated before queueing work; TUI reconnection recovery also handles offline /quit and /exit.
Suggested actions
- Upgrade Codex CLI to rust-v0.166.0, then set the session model to GPT-6.1 Terra from the bundled, Vertex AI, or Amazon Bedrock catalog.
- codex doctor --network-only
- After queueing a message onto an unloaded thread, run: codex queue --resume
- In the Codex TUI, enter keyboard transcript selection; move by word, line, and page; use integrated Find; set shortcuts if defaults conflict. Composer drafts stay intact while selecting.
- When a safety block appears, read the TUI explanation and pick a selectable safer follow-up prompt. Resume or fork the same thread—acknowledged approvals, permissions, and reasoning-token accounting are preserved.
Config surfaces this release may change
- background and headless runs — check your repo before applying
- sandbox settings — check your repo before applying
- skills — check your repo before applying
- hooks — check your repo before applying
- MCP servers — check your repo before applying
After you finish
Do not report this as applied to DevAgentRadar. You cannot write the visitor's loop.
Tell the human: open https://devagentradar.com/proposals/openai-codex-rust-v0-166-0-resume-unloaded-codex-threads-with-codex-queu and mark Applied, Skipped, or Failed. Proposal id: 1546988a-995f-4233-91e7-deb014cfd91d
Deliverables:
- Short impact assessment
- Optional patch plan (files + steps)
- Do not invent APIs not in the source notes
Your loop
This browser · no sign-in · not shared as “you”After you run the prompt
Only you can mark this. Agents cannot write your loop.
Your decision stays on this device. A public tally appears after a few votes.