Privacy Policy
Job Curator runs inside your own Google account. Your spreadsheet, your documents, and your service keys stay there. It has no server of its own, and we do not receive your documents, your spreadsheet, or your keys. What it sends to the outside services you choose is set out below.
Who we are
Job Curator is a Google Sheets tool made and operated by Paul Nicol, an individual based in Chicago, Illinois, United States. Contact: pnicol66@gmail.com.
What the tool accesses in your Google account
When you install Job Curator, Google's consent screen asks for six permissions. Some of them cover more than the tool uses. Here is what the tool does with each.
- Your spreadsheets. The tool works only in the spreadsheet it runs in. Its tabs hold the jobs you save, with a copy of each job description, your applications, a Usage log of each search and research request, and the tool's status notes. It does not open your other spreadsheets.
- Your Google Drive. Google grants this permission for your whole Drive. The tool works in a folder it creates named Job Copilot. There it keeps your career registry and evidence bank, any file you upload to be converted to a Google Doc, and a copy of your last search's listings. Each application gets its own folder: drafting files the cover letter and a notes document, and every CV build adds the tailored CV, PDFs of the CV and the letter, and a plain upload copy of the CV as a Google Doc, a PDF, and a Word file. Application folders move into Applied and Closed subfolders as their status changes. Outside that folder, the tool opens only the documents you register with it. Apart from finding its own Job Copilot folder by name, it does not browse or search the rest of your Drive, never changes who a file is shared with, and never deletes a file permanently: a file it replaces, or a draft you clear, goes to your Drive trash.
- Your Google Docs. The tool creates and edits the documents above. It reads the career registry and evidence bank you register, and adds a story to your evidence bank when you use Add SAR story. It reads your letters and CVs back when you review them, switch a letter's format, or build an upload copy. It does not open your other documents.
- Connecting to external services. Used for the job-search and research services you choose, and for Google's own Drive service, to convert an uploaded file to a Google Doc and to export the Word file. See "Third-party services" below.
- Running when you are not present. Used for the background runs you start (drafting and CV building on a timer, which remove themselves when finished) and for a trigger that responds when you change a status or tick a box in the tracker.
- Showing dialogs in the spreadsheet. Used for the tool's dialogs.
The tool also uses Google's map lookup (the Google Maps geocoder built into Apps Script) to measure distance for your location rule. It looks up place names only: your home base, any extra locations, and a listing's place when the job source sent no map position. Each place is looked up once and the result is kept with the spreadsheet.
Where your settings are kept. Your service keys and the identifiers of your registered documents are kept in Google's private per-user storage for this tool (Apps Script user properties). They are never written into spreadsheet cells, and the settings dialog never shows a saved key back to you. Your search criteria, your job title and industry lists, suggested titles, the map lookups, and the state of background runs are kept with the spreadsheet (Apps Script document properties).
The tool never reads your Gmail, calendar, contacts, or email address.
What we collect
No documents, spreadsheet contents, keys, or email addresses. Job Curator has no server of its own. Everything it does runs on Google's Apps Script platform inside your Google account, and no request passes through us.
One thing does reach us. Like every Apps Script add-on, the add-on's log lines go to the developer's Google Cloud project, where Paul reads them to fix problems. The tool writes them to say only what happened and how many: a fixed event name with counts, durations, HTTP status codes, token counts, the research model's name, and an error type. They never contain a job title, company, place, pay, criterion, research text, file or folder name, key, or anything from your documents. Google adds its own technical details, such as the time and the step that ran. Google's automatic error reports are turned off for the add-on, so an error message is shown to you and sent nowhere else.
If you use a spreadsheet copy of Job Curator instead of the add-on, its log lines are more detailed (job titles, companies, error messages) and stay in that copy's own script project, under the account that owns the copy.
If you write to us for support, we keep that email correspondence, and anything you attach such as a screenshot, like any other email.
Third-party services
To find jobs and to write your documents, the tool sends requests directly from your Google account to the services you choose, using keys you hold with them. What each receives:
- Job-search services (JSearch, through OpenWeb Ninja or RapidAPI, and Adzuna): the search terms (job titles, including any suggested titles you tick, and your own search terms); your home base, extra locations, or "remote"; the radius; the maximum posting age; and, for Adzuna, your pay floor when you exclude listings with no pay shown. Looking up one job's full description (when you approve a listing that came with only a preview, run Backfill missing job descriptions, or use the lookup in Add job by link) sends that job's title and company name with a location. Never your career registry, your stories, or your name.
- Your research provider (Anthropic for Claude, OpenAI, or Moonshot for Kimi, whichever you choose):
- Scoring, after each search and when you add a job by link: the first 3,500 characters of your career registry, which begins with your name and contact details; the first 6,000 characters of your evidence bank; your target titles and industries; your home base, radius, remote setting, pay target and floor, and seniority levels; and up to 40 listings at a time, each with a short excerpt.
- Each cover letter and each tailored CV: your full career registry, including your contact details; your full evidence bank; and the job's title, company, location, pay, fit notes, and up to 12,000 characters of its description.
- Review maybes: the first 8,000 characters of your registry and the first 12,000 of your evidence bank, your criteria including deal-breakers, and each parked job with its fit notes and an excerpt of up to 1,200 characters.
- Suggest more job titles from my stories, one request each time you click Suggest titles: your full career registry and evidence bank, the titles you already search, titles suggested before, and your seniority levels, employment types, target and excluded industries, and deal-breakers.
- Test connection: a one-line test message.
- Job-board and employer pages, in spreadsheet copies only. A spreadsheet copy of Job Curator opens the apply link of each scored listing, and of saved jobs when you run Check postings still open, to see whether the posting has closed. The request carries nothing of yours beyond the link itself. Indeed pages are never opened, and the add-on never opens these pages.
Those services handle that data under their own privacy policies and terms, which you accepted when you obtained their keys; the map lookup is Google's own service. Each key is sent only to the service it belongs to. Job Curator adds no service of its own to that path, and no request passes through us.
How Google user data is used and shared
Job Curator's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements. In particular, data obtained through Google APIs is used only to provide the features described on this page, is transferred only to the services you choose as described above, is not sold, is not used for advertising, and is not read by a human except with your permission for support, or as required by law.
Data retention and deletion
Everything the tool produces is in your own Drive and your own spreadsheet, and you control it. Deleting a file or the spreadsheet removes it, along with the settings kept with the spreadsheet. To stop using the tool: clear each stored key in API & research settings (type CLEAR in its box and save), use Stop background runs in the Applications menu, then remove Job Curator through your Google Account's third-party access settings or uninstall the add-on. That stops all access immediately; your files remain yours and are unaffected. What the tool sent to a job-search or research service is kept under that service's own policy. The developer's log lines are kept for Google Cloud Logging's standard retention period.
Children
Job Curator is a job-search tool for adults and is not directed at children under 16.
Changes
If this policy changes, the new version is published at this address with a new date.
Contact
Questions about this policy: pnicol66@gmail.com.