Cotera for Snowflake

Snowflake+Cotera

Run AI agents from a SQL query.

Your warehouse already defines the accounts, leads, and metrics that matter. With the Cotera Native App, a Snowflake query can call a predefined agent that reasons, acts, and returns its result, while Cotera handles execution, retries, and external systems.

Install from the Snowflake Marketplace. Authenticate with your Cotera API key.

When the work leaves the warehouse, agents reach the tools you already run

Salesforce
HubSpot
Slack
Notion
Linear
Attio
Reddit
Firecrawl

Capabilities

The warehouse selects the work, Cotera runs the agent.

SQL is already how your team expresses which rows matter. Cotera turns those selections into reliable agent workflows, not a one off LLM call that might succeed or fail silently.

Call agents from SQL

Invoke a predefined agent with cotera.agents.agent() or ask the default assistant with cotera.coco.chat(). Your query selects the rows, the agent does the work.

Predefined, governed agents

Agents are configured once in Cotera with their own tools, permissions, and instructions. SQL chooses which rows to send, not how the agent behaves.

Durable execution

Cotera handles retries, rate limits, and long running steps over HTTPS, so a call from a nightly job is reliable rather than a one off that might fail silently.

Work that leaves the warehouse

Create a CRM task, post to Slack, enrich from an external API, or open a support ticket. Cotera reaches the external systems your workflows depend on.

Complements Snowflake Cortex

Cortex handles in warehouse summarization, classification, and semantic queries. Cotera handles the multi step workflows that need to act outside the warehouse.

Observability and permissions

Every agent run is authenticated with your Cotera API key and observable in Cotera, so you can see what ran, what it did, and what it returned.

The pattern

Three steps, all in SQL.

Every Snowflake and Cotera workflow follows the same shape.

01

Select the work in SQL

Filter, join, and rank in Snowflake to build the list of accounts, leads, tickets, or cohorts that need an agent. Keep the prompts in a table or view so business logic stays versioned with your models.

02

Call the Cotera agent

Pass each row to cotera.agents.agent() with your agent ID. Cotera runs the agent over HTTPS, reasons across its tools, and returns the result straight to your query.

03

Store or trigger downstream

Write the returned result back to a table, join it to your models, or use it to trigger the next step. The output becomes warehouse data like anything else.

Better together

Cotera works alongside Snowflake Cortex.

They solve different halves of the problem. Cortex reasons over data inside the warehouse. Cotera runs the workflows that need to act outside it.

Snowflake Cortex

Snowflake Cortex

In warehouse summarization, classification, and semantic queries over your data. Ideal when the answer stays inside Snowflake.

Cotera

Cotera

Multi step workflows that leave the warehouse: CRM tasks, Slack posts, enrichments, support tickets, retries, rate limits, permissions, and observability across external systems.

Get started

Live in your warehouse in three steps.

01

Install from Marketplace

Get the Cotera Native App from the Snowflake Marketplace and install it into your account. Your admin approves outbound access to app.cotera.co during setup.

02

Connect your API key

Create a Cotera API key, store it as a Snowflake secret, and configure the app. The key inherits your organization access, so agents run with your permissions.

03

Run your first query

Copy any agent from the Solutions library, grab its agent ID, and call it from a worksheet. Snowflake selects the rows, Cotera returns the results.

WHAT ARE YOU WAITING FOR?

Ready to build?

Stop duct-taping tools.

Stop writing playbooks no one follows.

Start building agents that actually do the work—and do it well.

Cotera: For people who think busywork is boring.

"This is the best product ever when it comes to shipping. Ten on ten recommended. I just can't wait to see what happens with this product."

Rebecca Blount

Director, Customer Success