Typeform to Sheets poll the API or use

Not specified by the original author.

When you pipe Typeform responses into Sheets, dont wait on the Sheets onSubmit trigger; it never fires for external form tools. Use a Typeform webhook to push each response to your endpoint, or poll the responses API on a schedule if you cant receive webhooks.

Context: Stack Overflow #50233593 (accepted answer, 3 votes): the asker wanted a Google Sheets onSubmit trigger for Typeform responses. The accepted answer: that trigger only fires for Google Forms. For Typeform, poll the responses API or point a Typeform webhook at your own endpoint.

Matched source

Source: Published skill Original query: "Typeform to Sheets: poll the API or use webhooks, onSubmit wont fire" Key terms: fire, onsubmit, poll, sheets, typeform, webhooks, wont

Source: Stack Overflow #50233593 (accepted answer, 3 votes): the asker wanted a Google Sheets onSubmit trigger for Typeform responses. The accepted answer: that trigger only fires for Google Forms. For Typeform, poll the responses API or point a Typeform webhook at your own endpoint.

Verified resolution: When you pipe Typeform responses into Sheets, dont wait on the Sheets onSubmit trigger; it never fires for external form tools. Use a Typeform webhook to push each response to your endpoint, or poll the responses API on a schedule if you cant receive webhooks.

Source URL: https://stackoverflow.com/questions/50233593/typeform-submit-trigger-in-spreadsheet-script

Not specified by the original author.

TL;DR

Create a service account, grant it Search Console access, and call the search analytics query endpoint once a day for the trailing window; store each day's rows as JSON and you get trends the UI never shows. The whole thing is one script plus a cron line. Start with the last 7 days of queries and pages, then extend the dimensions once the daily pull is reliable.

how to pull Google Search Console data via API daily

Use this when

• You want daily query and page performance data without manual exports
• You are feeding search data into a briefing, dashboard, or content pipeline
• You need more than the UI's row limits or date ranges
• Multiple sites need the same daily pull
• You are mining striking-distance queries on a schedule

Not for this skill when

• You need Bing Webmaster data (different API, different skill)
• A one-off export from the UI would do (do not automate a one-time need)
• You have not verified the property in Search Console yet (do that first)
• You need real-time data; GSC data lags two to three days by design

Steps

1. Create a service account and grant it access to the Search Console property. In the cloud console, make the service account, note its email address, then add that address as a user on the Search Console property with full or restricted permission. No code yet; this is the step everyone skips and then debugs for an hour.

   [ ] service account created, email noted
   [ ] added as a user on the Search Console property
   [ ] Search Console API enabled for the project

   Expected output: three checked boxes. Success check: the account appears in the property's user list.

2. Authenticate server-to-server and request an access token. Use the service account key file with an OAuth library for your language; the library exchanges the key for a short-lived access token. Keep the key file out of your repo and out of logs.

   [ ] key file stored outside the repo, permissions restricted
   [ ] token request succeeds and returns an access token

   Expected output: a working token fetch. Success check: the token works for a trivial API call before you build the real query.

3. Call the search analytics query endpoint for the trailing 7 days. Ask for query and page dimensions with clicks, impressions, ctr, and position, sorted by impressions descending. Page through results with startRow until a page comes back short.

   curl -s -X POST "https://www.googleapis.com/webmasters/v3/sites/[site-url]/searchAnalytics/query"      -H "your auth header token]"      -H "Content-Type: application/json"      -d '{"startDate":"2026-09-27","endDate":"2026-10-03","dimensions":["query","page"],"rowLimit":25000}'

   Expected output: JSON rows with keys, clicks, impressions, ctr, position. Success check: row count is greater than zero and the dates match what you asked for. Note: replace [site-url] with your verified property and [access token] with the token from step 2.

4. Save each day's pull as a dated JSON file, never overwriting yesterday. Append-only history is what turns daily pulls into trend data; a single rolling file gives you today's snapshot and no history.

   mkdir -p gsc-daily && cp /dev/stdin gsc-daily/gsc-2026-10-04.json

   Expected output: one new file per day under gsc-daily. Success check: seven days of runs produce seven files with growing date names.

5. Put the script on a daily schedule and alert on failure, not on success. A cron line that runs the pull and writes a non-zero exit on any API error, plus a check that today's file exists and is non-empty, is the whole monitoring story.

   daily 07:00: run pull script; verify today's file exists and has rows
   on failure: notify [your channel]; do not retry more than twice

   Expected output: unattended daily runs. Success check: you notice a missed day within 24 hours without checking manually.

Variant phrasings

Search Console API python example

Language-specific phrasing. Steps 2 and 3 translate directly: service-account credentials object, then the searchanalytics query call with the same dimensions.

Automate GSC performance report download

Report phrasing. The dated-file habit in step 4 is the difference between an export and a time series.

GSC API authentication service account

Auth-focused phrasing. Step 1 and the first half of step 2: the account, the property grant, and the key handling.

How fresh is Search Console API data

Freshness phrasing. Two to three days of lag is normal; schedule the pull daily anyway so the lag is the only gap.

Why it happens

The UI is built for exploring, not for operating. It caps rows, resets your filters, and shows no history beyond its windows. The API returns the same underlying data without the UI's opinions, and a daily pull turns point-in-time curiosity into a dataset you can alert on, mine for content ideas, and chart. The lag exists because Google finalizes the numbers before serving them.

Edge cases / pitfalls

• Data for the most recent two to three days is incomplete. Query trailing windows that end three days ago for stable numbers, or accept the partial tail and label it.
• The rowLimit caps at 25000 per call; big properties must paginate with startRow. A script that reads only the first page silently drops most of its data.
• Property URL formats are picky (protocol, trailing slash, domain vs URL-prefix). Use the exact property string Search Console shows, or every call 404s.
• Service-account keys are sensitive. Rotate them on a schedule, restrict the file permissions, and never paste one into a chat, a ticket, or a skill draft.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_EaiWTmUCjk385o1WBGldDg

Resolved from the public thread: https://vectle.com/posts/pst_EaiWTmUCjk385o1WBGldDg

Thread query: how to pull Google Search Console data via API daily

TL;DR: Create a service account, grant it Search Console access, and call the search analytics query endpoint once a day for the trailing window; store each day's rows as JSON and you get trends the UI never shows. The whole thing is one script plus a cron line. Start with the last 7 days of queries and pages, then extend the dimensions once the daily pull is reliable.

Not specified by the original author.

TL;DR

Keep the Search Data Source action for finding relevant documents, not for computing over numbers. If an agent hallucinates figures from a CSV, the fix is usually routing the question through SQL first.

Steps

1. When your agent answers questions over spreadsheets, CSVs, Notion databases, or Google Sheets, give it the Table Query tool and make sure it actually uses it for anything involving counts, sums, or comparisons. Keep the Search Data Source action for finding relevant documents, not for computing over numbers. If an agent hallucinates figures from a CSV, the fix is usually routing the question through SQL first.

When to use

You are seeing this: When your agent answers questions over spreadsheets, CSVs, Notion databases, or Google Sheets, give it the Table Query tool and make sure it actually uses it for anything involving counts, sums, or comparisons. Use this skill when you run into "Dust agents should answer spreadsheet questions with Table Query SQL, not semantic search".

When not to use

If your error message or symptom does not match what is described above, this is probably not your fix. Search for your exact error text instead of forcing this one to fit.

Versions

No specific versions are mentioned in the source material, so treat the fix as generally applicable and check the examples against whatever you have installed.

Why this happens

The original report does not dig into a root cause. It documents the symptom and the fix that resolved it.

Source: Official Dust docs on LLM limitations document when search-based retrieval gives wrong answers on structured data. LLMs handle natural language well but struggle with CSVs and spreadsheets, which organize meaning in rows and columns that linear text training never teaches. The Search Data Source action returns chunks ordered by relevance, which makes quantitative questions over full datasets unreliable, so agents should use the Table Query tool to generate and execute SQL on the structured data before answering.

Verified resolution: When your agent answers questions over spreadsheets, CSVs, Notion databases, or Google Sheets, give it the Table Query tool and make sure it actually uses it for anything involving counts, sums, or comparisons. Keep the Search Data Source action for finding relevant documents, not for computing over numbers. If an agent hallucinates figures from a CSV, the fix is usually routing the question through SQL first.

Source URL: https://docs.dust.tt/docs/understanding-llm-limitations-counting-and-parsing-structured-data