Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Snowflake-Labs/snowflake-ai-kit --skill snowflake-cortexgit clone --depth 1 https://github.com/Snowflake-Labs/snowflake-ai-kitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/snowflake-labs/snowflake-ai-kit/snowflake-cortex)<a href="https://agentmods.dev/skills/snowflake-labs/snowflake-ai-kit/snowflake-cortex"><img src="https://agentmods.dev/badge/skills/snowflake-labs/snowflake-ai-kit/snowflake-cortex/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/snowflake-labs/snowflake-ai-kit/snowflake-cortex"><img src="https://agentmods.dev/badge/skills/snowflake-labs/snowflake-ai-kit/snowflake-cortex.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00037 | $0.00573 |
| Opus 5.5 | $0.00015 | $0.00229 |
| Sonnet 5.5 | $0.00007 | $0.00115 |
| Haiku 4.5 | $0.00004 | $0.00057 |
Grade A, and why
snowflake-cortex scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 16d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Snowflake Cortex AI
Use Snowflake's Cortex AI services when they are available on the MCP server.
When to use
- User asks a data question and semantic views are configured (Cortex Analyst)
- User wants to search unstructured data like support tickets or documents (Cortex Search)
- User wants to interact with a Cortex Agent for multi-step tasks
Discover available Cortex services
-- Semantic views (data models for Cortex Analyst)
SHOW SEMANTIC VIEWS IN ACCOUNT;
-- Cortex Search services
SHOW CORTEX SEARCH SERVICES IN ACCOUNT;
-- Cortex Agents
SHOW AGENTS IN ACCOUNT;
Cortex Analyst
Cortex Analyst converts natural language questions into SQL using semantic views. If the MCP server exposes a CORTEX_ANALYST_MESSAGE tool, pass the user's question as the message.
When to prefer Analyst over raw SQL:
- The question maps to metrics/dimensions defined in a semantic view
- The user is asking business questions ("what was revenue last quarter?")
- Governance and auditability matter
When to use raw SQL instead:
- Admin tasks (SHOW, DESCRIBE, GRANT)
- DDL/DML operations
- No semantic view covers the domain
Cortex Search
Cortex Search provides semantic search over unstructured text columns. If the MCP server exposes a CORTEX_SEARCH_SERVICE_QUERY tool:
{
"query": "billing dispute resolution",
"columns": ["TICKET_ID", "CUSTOMER_NAME", "REQUEST"],
"limit": 10
}
Filter syntax (optional):
{
"query": "network outage",
"filter": { "@eq": { "SERVICE_TYPE": "Cloud" } },
"limit": 5
}
Supported filter operators: @eq, @contains (arrays), @gte, @lte, composed with @and, @or, @not.
Cortex Agents
If the MCP server wraps a Cortex Agent (CORTEX_AGENT_RUN), send a text message. The agent decides which internal tools to use (SQL, Analyst, Search, code execution).
For multi-turn conversations, include thread_id from the previous response.
Examples
- "What was total revenue by region last quarter?" → Route to Cortex Analyst
- "Search support tickets about billing errors" → Route to Cortex Search
- "Analyze our customer churn data and create a visualization" → Route to Cortex Agent
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 16d ago First seen · 80 lines · 37 tokens per session scan A 340aea5d44fb
snowflake-cortex is a skill published in the GitHub repository Snowflake-Labs/snowflake-ai-kit (42 stars, last pushed 3d ago), licensed Apache-2.0. It adds 37 tokens to every session and 573 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-24.
snowflake-cortex compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| snowflake-cortexSnowflake-Labs/snowflake-ai-kit | Skill | 42 | 3d ago | Apache-2.0 | A | 37 | Claude Code |
| mapSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 42 | Claude Code |
| context-addSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 48 | Claude Code, Codex |
| cocoleanSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 41 | Claude Code |
| map-explainSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 26 | Claude Code |
| testSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 47 | Claude Code, Codex |
| map-diffSnowflake-Labs/cocoplus | Skill | 728 | 2d ago | MIT | A | 28 | Claude Code |