一键导入
handoff
Summarize the current conversation into a handoff file saved to ~/.claude/handoffs/
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Summarize the current conversation into a handoff file saved to ~/.claude/handoffs/
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Create native editable draw.io diagrams as .drawio files and optionally export them to PNG, SVG, PDF, JPG, or browser URLs. Use when Codex needs to create, generate, draw, or design a diagram, flowchart, architecture diagram, ER diagram, sequence diagram, class diagram, network diagram, mockup, wireframe, UI sketch, or work with draw.io, drawio, drawoi, .drawio files, or diagram export.
Automate browser interactions, test web pages and work with Playwright tests.
PyTorch model-building conventions and a neural-net training debug checklist. Use this skill whenever writing or reviewing PyTorch code that defines a model (nn.Linear, nn.Conv2d, BatchNorm) or trains one (training loop, optimizer, LR schedule), and ESPECIALLY when debugging training problems — loss not converging, loss stuck or flat, suspiciously slow learning, "hockey stick" loss curves, or a model that seems to ignore its inputs. Also consult it before starting any from-scratch training run, even if the user doesn't mention debugging.
| name | handoff |
| description | Summarize the current conversation into a handoff file saved to ~/.claude/handoffs/ |
date +%Y-%m-%ddate +%H%Mbasename $(pwd)pwdReview the entire conversation above and write a handoff file to:
~/.claude/handoffs/<project>_<date>_<time>.md
The handoff may be picked up by a different tool (Claude Code or Codex CLI), so keep it self-contained — no tool-specific assumptions.
Use this structure:
# Session Handoff — <date> <time>
**Project:** <project>
**From:** Codex CLI
**Working dir:** <full path>
## What we were doing
<1–3 sentences on the main task or goal>
## Decisions made
<bullet list of key decisions or conclusions reached>
## Current state
<where things stand — what's done, what's in progress, what's broken>
## Gotchas / don't redo
<dead ends already tried, constraints to respect, traps to avoid — so the next agent
doesn't repeat work you already ruled out>
## Next steps
<actionable list — enough detail to resume without re-reading the conversation>
## Key files / commands
<important file paths, commands, or snippets referenced in the session>
After writing the file, tell the user the exact path.