Build Your Own Coding Agent in 30 Minutes with Snowflake’s New code_toolset_all

Snowflake announced the General Availability of its Cortex Agents Coding Agent on August 26, 2026, enabling production-ready use of the code_toolset_all tool for building coding agents. A developer has created an open-source Python SDK and Streamlit chat UI that connect to the API, featuring keypair authentication, SSE streaming, and retry logic, with all code tested against a live Snowflake account.

TL;DR: Snowflake’s Cortex Agents now support code toolset all — a fully managed sandbox with bash, file I/O, SQL execution, and web search. This guide walks you through building a Python SDK client and a Streamlit chat UI that connects to it, complete with keypair auth, SSE streaming, and retry logic. All code is open-source and tested against a live Snowflake account. On August 26, 2026 , Snowflake announced the General Availability of the Cortex Agents Coding Agent. This isn’t a preview or beta — it’s production-ready, GA, and available to all accounts. The feature lets you add code toolset all to any Cortex Agent request, and Snowflake provisions a fully managed sandbox backed by the same runtime that powers CoCo Snowflake's built-in coding assistant . I wanted an AI agent that could: With GA status, this is now production-ready for real workloads. The catch? There was no Python SDK yet. So I built one. By the end of this article, you’ll have: Architecture Diagram You Streamlit → SDK → Cortex Agents API → Managed Sandbox → Response The Coding Agent API uses JWT tokens signed with your RSA private key. Generate one: openssl genrsa 2048 | openssl pkcs8 -topk8 -inform PEM \ -out ~/.ssh/snowflake rsa key.p8 -nocrypt openssl rsa -in ~/.ssh/snowflake rsa key.p8 -pubout \ -out ~/.ssh/snowflake rsa key.pub Register the public key in Snowflake: ALTER USER MY USER SET RSA PUBLIC KEY='MIIBIjANBgkq...'; The SDK has four core components: | Module | Responsibility || -------------- | -------------------------------- || config.py | Environment-based configuration || auth.py | JWT token generation and refresh || streaming.py | SSE event parsing || client.py | HTTP transport with retry | The Cortex Agents API streams responses using Server-Sent Events . Here’s what the actual wire format looks like: event: response.statusdata: {"message":"Provisioning sandbox","status":"sandbox provisioning"}event: response.text.deltadata: {"text":"Hello I'm","sequence number":2}event: response.text.deltadata: {"text":" Cortex Code.","sequence number":3}event: responsedata: {"content": {"type":"text","text":"Hello I'm Cortex Code."} ,"status":"completed"}event: donedata: DONE Most SSE parsers assume single-line data: events. The Cortex API uses two-line pairs event: then data: , which means you need to track state between lines: python def iter events self, response : current event type = Nonefor line in response.iter lines : if line.startswith "event: " : current event type = line 7: continue if line.startswith "data: " : data str = line 6: event type = current event type or "unknown" current event type = None if data str == " DONE ": yield StreamEvent event type="done" return data = json.loads data str yield StreamEvent event type=event type, data=data This was the 1 bug in my initial implementation — parsing each line independently produced empty responses. I wanted the SDK to feel natural for Python developers: python from cortex coding agent import ClientConfig, CodingAgentBuilderagent = CodingAgentBuilder .with model "claude-sonnet-4-5" .with system instructions "You are a data engineering assistant." .with workspace "USER$.PUBLIC.DEFAULT$" .with permission policy "always allow" .build ClientConfig with agent: response = agent.ask "Show me the top 5 tables by row count" print response.text The builder validates configuration at build time and constructs the proper API payload: { "messages": {"role": "user", "content": {"type": "text", "text": "..."} } , "tools": {"tool spec": {"type": "code toolset all", "name": "code toolset all"}} , "models": {"orchestration": "claude-sonnet-4-5"}, "tool resources": { "code toolset all": { "permission policy": {"type": "always allow"}, "workspace mounts": {"name": "USER$.PUBLIC.DEFAULT$", "mount path": "/workspace"} } }} Here’s a subtlety that cost me debugging time: the API manages conversation state server-side . You don’t send message history — you send only the latest message, and the API maintains continuity via thread id. My first implementation accumulated all messages client-side and sent the full history. This caused the API to return empty responses on the 2nd and 3rd turns. The fix: php def build request self, user message: str - AgentRequest: message = Message role=MessageRole.USER, content= ... Send ONLY the current message — API manages history return AgentRequest messages= message , thread id=self. thread id, Server tracks state ... The entire chat interface is remarkably simple once the SDK handles the complexity: python import streamlit as stfrom cortex coding agent import ClientConfig, CodingAgentBuilder@st.cache resourcedef get agent : return CodingAgentBuilder .with model "claude-sonnet-4-5" .with permission policy "always allow" .build ClientConfig if prompt := st.chat input "Ask the coding agent..." : with st.spinner "Thinking..." : reply = get agent .ask prompt .text st.markdown reply Launch with: export SNOWFLAKE ACCOUNT=myaccount.region.cloudexport SNOWFLAKE USER=MY USERexport SNOWFLAKE PRIVATE KEY PATH=~/.ssh/snowflake rsa key.p8streamlit run examples/streamlit app.py Once running, the agent handles prompts like: | Prompt | What Happens || --------------------------------------- | -------------------------------------- || “Show me all databases” | Executes SHOW DATABASES via SQL tool || “Write a Python script to parse JSON” | Generates code in the sandbox || “Read the CSV at /workspace/data.csv ” | Uses file read tool || “Search for pandas documentation” | Uses web search tool || “Create a stored procedure for ETL” | Combines SQL + code generation | The sandbox includes numpy, pandas, scipy, matplotlib, and plotly pre-installed. Need more? Add artifact repositories for PyPI access. 1. SSE parsing is harder than it looks. The two-line event/data format is standard SSE, but most tutorials show single-line examples. Always pair event: with the next data: line. 2. Don’t accumulate message history. The Cortex API is stateful — it tracks conversations server-side. Sending history causes silent failures. 3. Use explicit Python paths on macOS. If you have Anaconda installed, /usr/bin/python3 -m venv .venv avoids path conflicts that cause mysterious PermissionError failures. 4. Pin cryptography<44 for Python 3.9 compatibility on macOS. Newer versions have Rust bindings that don't work with the system OpenSSL. 5. The sandbox provisions on first call ~10s then subsequent calls in the same thread are near-instant. Design your UX around this cold-start pattern. Before deploying to your team: The complete SDK 38 tests, Streamlit app, deployment SQL is available as a single zip: Repository structure: cortex-coding-agent/├── src/cortex coding agent/ SDK 10 modules ├── tests/unit/ 38 passing tests├── examples/streamlit app.py Chat UI├── deploy/snowflake/ Agent DDL + RBAC└── docs/ Install guide + extension docs The Cortex Agents Coding Agent code toolset all fundamentally changes how you build AI-powered applications on Snowflake. Instead of managing agent loops, sandboxes, and tool infrastructure yourself, you make a single API call and Snowflake handles the rest. In this article, we built a complete Python SDK from scratch — keypair authentication, SSE streaming, a fluent builder, and a Streamlit chat UI — all validated against a live Snowflake account with 38 passing tests. The key takeaway: the barrier to building a production coding agent is now one REST endpoint and a few hours of work. No servers, no containers, no orchestration framework. Just your application, a keypair, and a POST request. Since code toolset all automatically inherits new tools as Snowflake adds them, your SDK stays current without code changes. That's the real power of a managed runtime — you focus on the application, not the infrastructure. Found this useful? A 👏 helps others find it too. Honored to be a finalist for the 2026 Snowflake Community Awards — Open Source Impact APJ If this project helped you, I’d love your vote: 🎥 1-min video: How to vote https://app.snowflake.com/static/workspaces/workspaces-wave1.html?v=1787904396586 🗳️ Vote here https://app.snowflake.com/static/workspaces/workspaces-wave1.html?v=1787904396586 Page 6 → APJ → Satish Kumar | LTIMindtree, India Voting closes Sep 15 . Your support means a lot 💙 Follow for weekly Snowflake engineering deep dives: You may use, share, adapt, and build upon this work. For public redistribution or substantial adaptation, please retain attribution and include a link to the original article or repository and the author’s LinkedIn profile. Private and internal use requires no attribution. Provided “as is” for educational purposes. Please validate and test all examples before using them in production. Views are my own and do not represent any current or former employer. Build Your Own Coding Agent in 30 Minutes with Snowflake’s New code toolset all https://pub.towardsai.net/build-your-own-coding-agent-in-30-minutes-with-snowflakes-new-code-toolset-all-e05241d0c572 was originally published in Towards AI https://pub.towardsai.net on Medium, where people are continuing the conversation by highlighting and responding to this story.