Snowflake Intelligence

Snowflake Intelligence is the supported agent integration path for this version of Metatate.

Metatate keeps its canonical governance tools and MCP server intact:

core.metatate_mcp
core.discover_context(input VARIANT) and peers

But Snowflake Intelligence custom tools bind named scalar parameters, not a single VARIANT request object. To bridge that mismatch, Metatate provides an adapter layer of Snowflake Intelligence-friendly wrappers:

core.agent_discover_context
core.agent_get_decision_context
core.agent_inspect_data_meaning
core.agent_authorize_use
core.agent_validate_query_context
core.agent_explain_why

These wrappers forward into the canonical Metatate tools and return the same JSON payloads.

Prerequisites

Metatate installed and running in your account
At least one policy deployed so governance tables contain data
Snowflake Intelligence enabled for your account and role
An Intelligence agent created in a schema where your role can manage agents

Step 1: Create a Snowflake Intelligence agent

Open AI & ML -> Agents in Snowsight
Click Create agent
Choose a writable database and schema
Create an agent such as:
- object name: METATATE_TEST_AGENT
- display name: Metatate Test Agent

Step 2: Add the Metatate custom tools

Open the agent's Tools tab and add the following six custom tools.

1. `core.agent_discover_context`

Resource type: function
Object: core.agent_discover_context
Parameters:
- database_name -- string
- schema_name -- string
- min_sensitivity -- string
- compliance_framework -- string
- has_pii -- boolean
- domain -- string

Use this for questions like:

Which confidential tables in DEV contain PII?

2. `core.agent_get_decision_context`

Resource type: function
Object: core.agent_get_decision_context
Parameters:
- table_name -- string, required

Use this for questions like:

What governance policies apply to DEV.PUBLIC.LINEITEM?

3. `core.agent_inspect_data_meaning`

Resource type: function
Object: core.agent_inspect_data_meaning
Parameters:
- table_name -- string, required
- column_name -- string

Use this for questions like:

What does the EMAIL column mean from a governance perspective?

4. `core.agent_authorize_use`

Resource type: procedure
Object: core.agent_authorize_use
Parameters:
- table_name -- string, required
- operation -- string, required
- intended_use -- string, required
- actor_role -- string
- destination_system -- string
- destination_jurisdiction -- string

Use this for questions like:

Can I export DEV.PUBLIC.LINEITEM to Salesforce for reporting?

5. `core.agent_validate_query_context`

Resource type: procedure
Object: core.agent_validate_query_context
Parameters:
- sql_text -- string, required
- operation -- string
- intended_use -- string
- actor_role -- string

Use this for questions like:

Validate this query: SELECT L_ORDERKEY FROM DEV.PUBLIC.LINEITEM

6. `core.agent_explain_why`

Resource type: procedure
Object: core.agent_explain_why
Parameters:
- decision_id -- string, required

Use this for questions like:

Why was decision dec_abc123 denied?

Step 3: Save the agent

Save the agent after adding or editing the custom tools. Snowflake Intelligence chat threads do not reliably pick up unsaved tool changes.

Step 4: Test the first workflow

Start a new thread with the agent and ask:

What governance policies apply to DEV.PUBLIC.LINEITEM?

Expected behavior:

the agent calls core.agent_get_decision_context
Metatate returns the standard decision-context JSON payload
the agent summarizes policy, sensitivity, instructions, compliance, and risk metadata

Practical prompt examples

Intent	Prompt	Expected wrapper
Discovery	`Which confidential tables in DEV contain PII?`	`core.agent_discover_context`
Decision context	`What governance policies apply to DEV.PUBLIC.LINEITEM?`	`core.agent_get_decision_context`
Meaning inspection	`Show me the governance meaning of the EMAIL column in DEV.PUBLIC.CUSTOMERS.`	`core.agent_inspect_data_meaning`
Authorization	`Can I export DEV.PUBLIC.LINEITEM to Salesforce for reporting?`	`core.agent_authorize_use`
Query validation	`Validate this query: SELECT L_ORDERKEY FROM DEV.PUBLIC.LINEITEM`	`core.agent_validate_query_context`
Decision explanation	`Why was decision dec_abc123 denied?`	`core.agent_explain_why`

Troubleshooting

The agent says it hit a SQL compilation error

Most likely cause: you wired a canonical core.*(input VARIANT) tool directly into the agent instead of the corresponding core.agent_* wrapper.

The agent ignores a newly added tool

Save the agent, then start a new thread. Unsaved changes and old threads may not use the latest tool configuration.