Save team slash commands in Codex `/presets` slots
Codex CLI rust-v0.164.0 adds ten numbered/presets slots with a keyboard menu, plus inline /model and /permissions arguments. Capture repeatable prompts as presets and switch model, reasoning, speed, and permission profiles without leaving the session.Why this loop
This release makes Codex CLI an in-session control plane instead of a restart-and-retype loop. Reusable prompts and slash commands can live in ten numbered /presets slots with a keyboard menu, so review, test, and MCP playbooks are one keystroke instead of paste from docs. Inline /model arguments set model, reasoning effort, and speed without editing files; inline /permissions arguments apply permission profiles in place. Queued prompts now wait for pending permission selections, so preset bursts will not race ahead of approvals. Enterprise MCP can authenticate with eligible Codex account grants or dedicated sign-in, including OAuth clients that require secrets. Credentials are masked during supported remote Bash commands and Zsh snapshot sessions, including custom providers. Compaction failures no longer leave partial summaries, and resume finds compressed histories and restores saved thread settings instead of global defaults. Encode the team's playbooks as presets and switch model/permission profiles inline.
Proposed actions
- Upgrade to OpenAI Codex CLI rust-v0.164.0, then run
/presetsand save the team's repeatable prompts and slash commands into the ten numbered slots via the keyboard menu. - Mid-session, set model, reasoning effort, and speed with inline
/modelarguments instead of editing config or restarting Codex. - Apply a permission profile with inline
/permissionsarguments before sending queued prompts so those prompts wait until the pending permission selection resolves. - Authenticate enterprise MCP connections with eligible Codex account grants or dedicated MCP sign-in, including OAuth clients that require secrets; do not paste client secrets into session config.
- On Kitty or Sixel terminals, preview viewed and generated images in the fullscreen transcript instead of exporting them to an external viewer.
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: Save team slash commands in Codex /presets slots
Summary: Codex CLI rust-v0.164.0 adds ten numbered /presets slots with a keyboard menu, plus inline /model and /permissions arguments. Capture repeatable prompts as presets and switch model, reasoning, speed, and permission profiles without leaving the session.
Primary source: https://github.com/openai/codex/releases/tag/rust-v0.164.0
Why it matters
This release makes Codex CLI an in-session control plane instead of a restart-and-retype loop. Reusable prompts and slash commands can live in ten numbered /presets slots with a keyboard menu, so review, test, and MCP playbooks are one keystroke instead of paste from docs. Inline /model arguments set model, reasoning effort, and speed without editing files; inline /permissions arguments apply permission profiles in place. Queued prompts now wait for pending permission selections, so preset bursts will not race ahead of approvals. Enterprise MCP can authenticate with eligible Codex account grants or dedicated sign-in, including OAuth clients that require secrets. Credentials are masked during supported remote Bash commands and Zsh snapshot sessions, including custom providers. Compaction failures no longer leave partial summaries, and resume finds compressed histories and restores saved thread settings instead of global defaults. Encode the team's playbooks as presets and switch model/permission profiles inline.
Suggested actions
- Upgrade to OpenAI Codex CLI rust-v0.164.0, then run
/presetsand save the team's repeatable prompts and slash commands into the ten numbered slots via the keyboard menu. - Mid-session, set model, reasoning effort, and speed with inline
/modelarguments instead of editing config or restarting Codex. - Apply a permission profile with inline
/permissionsarguments before sending queued prompts so those prompts wait until the pending permission selection resolves. - Authenticate enterprise MCP connections with eligible Codex account grants or dedicated MCP sign-in, including OAuth clients that require secrets; do not paste client secrets into session config.
- On Kitty or Sixel terminals, preview viewed and generated images in the fullscreen transcript instead of exporting them to an external viewer.
Config surfaces this release may change
- slash commands — check your repo before applying
- MCP servers — check your repo before applying
- permission rules — check your repo before applying
- skills — 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-v0-164-0-save-team-slash-commands-in-codex-presets-slo and mark Applied, Skipped, or Failed. Proposal id: 41d02d2c-b27d-4e00-b48f-1a7d4f5d0f46
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.