메뉴
Inspect and safely maintain local Codex App/CLI state with report-first, backup-first, archive-only workflows for sessions, logs, worktrees, config project entries, and handoff reminders.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | keep-codex-fast |
| description | Inspect and safely maintain local Codex App/CLI state with report-first, backup-first, archive-only workflows for sessions, logs, worktrees, config project entries, and handoff reminders. |
Use this skill when Codex feels slow, local session history has grown, or the user asks for safe Codex App/CLI local-state maintenance.
config.toml.--details unless the user asks for raw thread IDs, titles, local paths, or process paths.--apply while Codex is active unless the user explicitly asks to wait for Codex to exit.--apply, archive, prune, rotate, normalize, delete, or mutate local state automatically.Run the bundled helper through the skill path:
python3 ~/.codex/skills/keep-codex-fast/scripts/keep_codex_fast.py
Summarize the report:
If the user wants a safer pre-apply checkpoint, run backups only:
python3 ~/.codex/skills/keep-codex-fast/scripts/keep_codex_fast.py --backup-only
If the user explicitly asks to apply maintenance, first confirm that important active chats have handoffs or are safe to archive. Then ask them to close Codex, or run with --wait-for-codex-exit if they accept waiting:
python3 ~/.codex/skills/keep-codex-fast/scripts/keep_codex_fast.py --apply --archive-older-than-days 10 --worktree-older-than-days 7 --wait-for-codex-exit
Verify with another report-only run afterward:
python3 ~/.codex/skills/keep-codex-fast/scripts/keep_codex_fast.py
When the user needs continuity before archiving a chat, use references/handoff-template.md.
Prefer repo-local handoff paths such as:
docs/codex-handoffs/YYYY-MM-DD-topic.md
When the user asks for recurring maintenance, create only a reminder/report automation. The automation prompt should run the report first, summarize the counts, remind the user to create handoffs, and explicitly avoid --apply.
Improve, audit, simplify, rewrite, or migrate prompts, tool descriptions, agent instructions, and prompt stacks for GPT-5.6 Sol or the GPT-5.6 family. Use for outcome-first prompt design, autonomy boundaries, tool routing, PTC, grounding, verbosity, reasoning effort, and prompt evals.
Apply a deep, design-minded engineering workflow for complex work that needs careful planning, iteration, and simplification.
Use when preparing or running a Claude CLI agentic engineering review through claude -p for repository audits, architecture analysis, system design, security review, large refactors, specs, RFCs, or evidence-grounded long-form engineering analysis.
Use when preparing or running a ZCode GLM-5.2 agentic builder workflow through zcode --prompt for fast implementation, iterative code generation, focused fixes, and validation on an existing repository.
Apply adversarial opposite-analysis to plans, specs, architecture, code changes, and claims. Use when the user asks for adversarial review, opposing analysis, contrarian review, red-team reasoning, or Z.ai and MiniMax cross-checks through the Ralph MCP router.
Review and adjudicate Bugbot, Cursor, Seer, and similar automated PR feedback with local evidence before accepting, fixing, or dismissing findings.