Nightly site updates as drafts
A nightly job updates your website content from a data export but can only save drafts. A check flags odd changes, and a person publishes only the safe ones.
What it does
- Reads the nightly export of prices and availability
- Writes changes to the site as drafts only
- Compares drafts with the live pages
- Flags big price moves or too many changes at once
- Lists the drafts that are safe to publish
Your approval boundaries
Approval requiredThese actions always wait for your yes:
- Publishing any draft
You decide what happens next.
More about this workflow
This setup is for operators and managers who keep the busywork moving. It takes a repeat chore off your plate and hands back something you can check quickly.
First, it reads the nightly export of prices and availability. Next, it writes changes to the site as drafts only. Then it compares drafts with the live pages. After that, it flags big price moves or too many changes at once. Finally, it lists the drafts that are safe to publish. It works with your database and spreadsheets, each on its own login with limited access.
You get the prep work done and waiting for you. The agent stops and asks for your yes before publishing any draft, and says plainly if a step cannot be undone.
“a token that lets it write drafts but never publish”
Explore required tools and model recommendations
Tools and access
Supabase
Handles the database part of this job. Grant a role limited to one table.
API required: Yes. Needs a Supabase project and an API key. Give the agent a role limited to one table.
Google Sheets
Handles the spreadsheets part of this job. Grant access to the single sheet it needs.
API required: Yes. Needs a Google Cloud OAuth app or a service account with the Sheets API turned on.
Official links only. Not sponsored, no affiliate links.
Model recommendations
Best pick
GPT-6.1 Sol
Cheap cached input suits jobs that rerun the same instructions each week.
$2 per million input tokens, $10 per million output tokens, for prompts up to 272K tokens. Longer prompts cost $4 in and $15 out.
OpenAI docs as of October 11, 2026
Cheaper backup
Gemini 3.5 Flash-Lite
Low cost for simple data pulls and matching.
$0.30 per million input tokens, $2.50 per million output tokens.
Google docs as of October 11, 2026
Our own picks. Not sponsored, no affiliate links.
Build your prompt
Add your details, then copy the prompt into your tool.
2Your prompt
4 of 4 still emptyYour answers stay in this browser only.
Run your workflow
Connect your tools with only the access this setup needs, then follow these steps.
Steps
Do these in order. Test on old data before it touches live work.
- 1Connect Supabase with its own login, not yours. Grant a role limited to one table.
- 2Connect Google Sheets with its own login, not yours. Grant access to the single sheet it needs.
- 3Paste the prompt where your harness takes instructions. See the note under the Copy button.
- 4Do a dry run on old or test data first. Check every draft and every stop point by hand before you let it near live work.
- 5Read its log after the first real run. Only widen access once a week of runs looks clean.
Safeguards and trust score
The score describes this example setup’s access, approval boundaries, ability to undo changes, and scope.
Can draft things that send or spend, so it sits close to risky actions.
Waits for a yes on: publishing any draft.
Leaves drafts you can throw away, with a record of what it made.
Runs on its own login across 2 tools. Fewer connections means a smaller blast radius.
How to raise it
- Access: Take away its ability to send, post or spend. Let it only write drafts, and have a person press send or pay from their own account +6
- Human approval: Add a second named stop next to "publishing any draft", so every step that leaves your hands waits for a yes, and have the confirm say plainly what cannot be undone +3
- Scope: Run it on one tool only, with its own login that can see just what this job needs +2
Do all of these to reach 91, up from 80.
Want help getting there? Get an AI Trust Audit from Pixeldesigns
Reviewed by Trustcase (example setup). How scoring works
Embed badge
Show this trust score on your site or README. It links back here.
[](https://trustcase.dev/use-cases/nightly-site-updates-as-drafts)<a href="https://trustcase.dev/use-cases/nightly-site-updates-as-drafts"><img src="https://trustcase.dev/badge/nightly-site-updates-as-drafts.svg" alt="Trustcase trust score for Nightly site updates as drafts" height="20"></a>More operations workflows
- Weekly brand priorities with review gate
A small set of named agents runs a small brand's week. One sets priorities, one drafts ads, one checks strategy and one does the numbers, and a different agent reviews each draft before a person approves it.
- Weekly KPI summary with number checks
Every Monday an agent writes the short management summary on supplier and delivery performance. It only sees data through read only tools, a control chart sets the traffic light, code checks every number and a person approves before it goes out.
- Founder morning stats and follow up brief
A scheduled agent with your tools connected sends a morning brief: yesterday's numbers, what happened in the product and today's action items. It remembers what you said you would do and follows up.
- Invoice inbox
Reads bills that land in email and gets them ready to pay.