Snowflake Metrics Digest with Codex
Execution here is asynchronous: submit a statement, take the handle back, then collect the result with that handle, polling anything still running rather than sending it again. That is API state handling more than analysis, and it is the kind of concrete sequencing Codex is comfortable with across five statements in a row.
Watch it work before it's live
Run a staged conversation — no account needed. The agent handles it for real while a simulated world answers its tool calls; nothing touches real accounts, and nothing is actually sent.
Morning metrics 12 August
The warehouse resumed in four seconds and all five statements finished inside the same minute. Gross merchandise value came back at 812,440.20 MXN against 774,910.05 yesterday, orders at 3,188 against 3,050, returns at 214, and active stores at 61 with no change.
Set up in minutes
Using this template drops you into a guided setup. It asks exactly this, nothing else:
Connect Snowflake
One sign-in. The agent acts through your account, scoped to what this template uses.
Connect Slack
One sign-in. The agent acts through your account, scoped to what this template uses.
Metrics queries
The read only statements to run each morning, each with the name it should carry in the digest and its database and schema written out in full. Nothing outside this list is ever submitted to your warehouse.
Digest style
What the digest should look like when it lands - which metric leads, the movements worth flagging (weekly active accounts under 4,000, on time delivery under 90 percent), and who gets tagged when a statement fails.
Runs on Codex
Preselected for this page — connect your Codex account during setup, or switch to NoClick's built-in models with one click.
Watch it handle a test run
A staged conversation against a simulated world — then it’s live.
Why Codex for this agent
Poll the handle, never resubmit
A statement still executing is asked again, not sent again. Resubmitting runs it twice and bills it twice, which is the mistake this pattern exists to prevent.
Five statements, one message
Submissions, collections and a single Slack post, in order, with no branching. The chain is long but entirely defined, which suits a fast harness.
Before you fork
What happens if a statement is still running when the digest is due?
The digest goes out without it and the statement is listed under FAILED STATEMENTS with what is known. That is deliberate, since a late digest is worse than an incomplete one for people starting their day. A statement that regularly runs long belongs in a materialised view rather than in the list.
Run it with a different agent
Put Snowflake Metrics Digest to work on Codex
Free to start. Guided setup, a test run against staged conversations, and it's live.