原文核对：2026-10-13 · 中文正文 · 英文提示词与演示
场景定位
规范输入，运行获准脚本或 API，对有限范围失败重试，以日志或产物验证后报告。
- 难度
- 中级
- 时间跨度
- 30 分钟
适合用于
- 输入结构化、授权明确且要求可审计结果的运营任务。
- 权限、批量邀请、配额、客户设置、路由检查和迁移跟进等重复工作。
- 需要准确报告成功、失败和待人工决定项的团队。
起步提示词
英文原文 · 可复制后按实际材料修改
I need to run this workflow:
Goal: [what should happen]
Inputs: [CSV, Google Sheet, list, ticket, or file path]
Approval or policy source: [Slack thread, doc, ticket, or none]
Runner: [script, API, CLI, skill, or manual app workflow]
Verification artifact: [result CSV, log, dashboard, screenshot, or other proof]
Please:
- inspect the inputs and ask only for missing required fields
- normalize dates, amounts, owners, and IDs before running the workflow
- run a dry run first when the workflow supports it
- run only the approved scope
- record one success or failure row per item
- retry transient failures once without restarting successful rows
- summarize totals, failures, retries, and verification artifacts
Pause before irreversible actions or scope changes.执行可审计操作
授权用户、批量更新、以不同参数运行脚本等周期工作，可以自动化并生成审计输出。
适用于重复执行并需提供验证产物的操作。
描述任务和输入
- 提供要批处理的表格、文件、工单或清单。
- 适用时指出授权来源或范围政策。
- 指定脚本、API、技能、CLI 或应用流程。
- 支持时先要求 dry run。
- 执行批处理，为每项记录一行成功或失败。
限定范围，所需输入齐全才执行；缺必填字段的行应标记，不能猜测。
通过插件连接工单系统或待办表格等工具。
提供结果证据
交付可检查的 CSV、日志、仪表盘链接、截图、PR 检查等证明；桌面应用中可打开生成文件核验。
保存为可复用流程
首轮成功后，把重复部分封装为技能或定时任务。
官方提示词
英文原文 · 可复制后按实际材料修改
Turn this operations run into a [reusable ChatGPT skill/scheduled task].
Capture:
- required inputs
- approval or policy source
- runner command, API, skill, or app workflow
- dry-run behavior if applicable
- verification artifact
- retry policy
- final report format
Keep the operation narrow and pause before irreversible actions.只有手动输出可靠后才安排定时执行；影响权限或数据的敏感操作默认持续只生成草稿，除非你明确要求执行。
展开英文原文对照
Use ChatGPT to normalize inputs, run approved scripts or APIs, retry bounded failures, and verify the result from logs or artifacts before reporting back.
Run operations you can audit
If you have repeatable operations you need to run regularly, such as giving access to a user, applying a batch update, or calling a script with different parameters for example, you can use ChatGPT to automate it and give you an auditable output.
Use this workflow when ChatGPT should run a repeatable operation and show you what happened with an artifact that counts as verification.
Describe the task and inputs
- Give ChatGPT the input table, files, tickets, or other list it should batch run the process on.
- Point it to the approval source or policy that defines the allowed scope, if applicable.
- Tell ChatGPT which script, API, skill, CLI, or app workflow should do the work.
- Optionally, ask for a dry run when the workflow supports one.
- Ask ChatGPT to run the batch operation and record one success or failure row per item.
Keep the scope narrow, and add instructions for ChatGPT to run the operation only when it has all the required inputs. If a row is missing a required field, ChatGPT should flag that row instead of guessing.
Connect the tools you use to run the operation with plugins, for example your ticketing system or your spreadsheet with list items.
Require proof to verify the result
A useful operations run includes a result that you or a teammate can inspect, such as a CSV, log file, dashboard link, screenshot, PR check, or other proof that the operation was successful. In the ChatGPT desktop app, you can open and inspect generated files after the run to verify the result.
Turn the run into a reusable workflow
After the first successful run, ask ChatGPT to capture the repeatable parts. For common workflows, this can become a skill or a scheduled task.
官方提示词
英文原文 · 可复制后按实际材料修改
Turn this operations run into a [reusable ChatGPT skill/scheduled task].
Capture:
- required inputs
- approval or policy source
- runner command, API, skill, or app workflow
- dry-run behavior if applicable
- verification artifact
- retry policy
- final report format
Keep the operation narrow and pause before irreversible actions.For scheduled operations, create a scheduled task only after the manual run produces reliable output. Keep sensitive actions that might affect access or data permanently draft-only unless you explicitly want ChatGPT to take them.