Replace <output_efficiency> with <response_guidelines> in Grok CLI prompts
Grok CLI renames the system prompt tag <output_efficiency> to <response_guidelines>, strips trailing /** on sandbox allow paths, restricts session dirs to owner-only, and installs native arm64 on Apple Silicon (including Rosetta/x86_64 updaters).
Why this loop
This release changes prompt structure and sandbox/session path handling. Custom prompts still using <output_efficiency> will not match the new <response_guidelines> contract. Sandbox allow entries ending in /** are stripped, so path rules may broaden or stop matching as intended. Session directories are now owner-only, which can break shared or multi-user layouts. Apple Silicon users still on Rosetta/x86_64 builds should move to the native arm64 install path. Align local configs and install method with these monorepo sync changes before relying on agent runs, pager tooling, or automations cards.
Proposed actions
- rg -n '<output_efficiency>|output_efficiency' ~/.grok . --glob '!**/node_modules/**' and rewrite every system-prompt hit to <response_guidelines>; keep surrounding instructions intact
- List every Grok/CLI sandbox allow path in your config and delete a trailing /** from each entry (release strips trailing /** on sandbox allow paths)
- stat -f '%Sp %N' <your-grok-session-dirs> and chmod 700 any session directory that is not owner-only (release restricts session directories to owner-only permissions)
- On Apple Silicon: from a Rosetta or x86_64 shell, reinstall/update so the CLI installs native arm64 (supported including from Rosetta shells and x86_64 updaters)
- If an alpha channel update fails, reinstall using GROK_CHANNEL as the release directs for alpha update failures
Agent prompt
Paste into your agent or query via MCP (`get_agent_prompt`) — free, no extra AI costPaste into Claude Code / CLAUDE.md task
# DevAgentRadar → Claude Code You are helping me adopt a real coding-assistant change. Work only from the facts below. Do not invent features. ## Context Assistant: xAI Grok CLI (Grok Build) Proposal: Replace <output_efficiency> with <response_guidelines> in Grok CLI prompts Summary: Grok CLI renames the system prompt tag <output_efficiency> to <response_guidelines>, strips trailing /** on sandbox allow paths, restricts session dirs to owner-only, and installs native arm64 on Apple Silicon (including Rosetta/x86_64 updaters). Primary source: https://github.com/xai-org/grok-build/commit/be713136d2a69080743a3f6b3c72077057e5948f ## Why it matters This release changes prompt structure and sandbox/session path handling. Custom prompts still using <output_efficiency> will not match the new <response_guidelines> contract. Sandbox allow entries ending in /** are stripped, so path rules may broaden or stop matching as intended. Session directories are now owner-only, which can break shared or multi-user layouts. Apple Silicon users still on Rosetta/x86_64 builds should move to the native arm64 install path. Align local configs and install method with these monorepo sync changes before relying on agent runs, pager tooling, or automations cards. ## Suggested actions 1. rg -n '<output_efficiency>|output_efficiency' ~/.grok . --glob '!**/node_modules/**' and rewrite every system-prompt hit to <response_guidelines>; keep surrounding instructions intact 2. List every Grok/CLI sandbox allow path in your config and delete a trailing /** from each entry (release strips trailing /** on sandbox allow paths) 3. stat -f '%Sp %N' <your-grok-session-dirs> and chmod 700 any session directory that is not owner-only (release restricts session directories to owner-only permissions) 4. On Apple Silicon: from a Rosetta or x86_64 shell, reinstall/update so the CLI installs native arm64 (supported including from Rosetta shells and x86_64 updaters) 5. If an alpha channel update fails, reinstall using GROK_CHANNEL as the release directs for alpha update failures ## Config surfaces this release may change - sandbox settings — check your repo before applying ## Your job 1. Restate the change in one sentence. 2. Propose a minimal plan for my repo (or a throwaway pilot). 3. Implement only what I approve; prefer small diffs and tests. 4. Call out risks (permissions, breaking APIs, cost). Start by confirming you understood the proposal.
Your loop
This browser · no sign-in · not shared as “you”Your decision stays on this device. A public tally appears after a few votes.