talonic_list_agent_tools
List the platform's agent tool registry: every retrieval, provenance and analysis primitive the in-product Talonic agent runs on (find_data, describe_data, query_data for read-only SQL over the extracted data, get_document_markdown, workspace_overview, …), each with its JSON input schema and whether this credential may invoke it. An API key reaches the read-only subset; the platform enforces the capability matrix on every call.
When to use
- You need a capability the dedicated
talonic_*tools do not cover (SQL over structured data, a workspace overview, cohort discovery). - You want to see exactly what the in-product agent can do before driving it yourself with
talonic_invoke_agent_tool.
When not to use
- Anything a dedicated
talonic_*tool already does — prefer those, they are shaped for you.
Parameters
| Parameter | Type | Description |
|---|---|---|
| only_invocable | boolean | Hide tools this credential cannot invoke. Default true. |
| include_schemas | boolean | Include each tool's JSON input schema. Default true. |
Response shape
Fields
| Parameter | Type | Description |
|---|---|---|
| tools[].name | string | Tool name to pass to `talonic_invoke_agent_tool`. |
| tools[].can_invoke | boolean | Whether this credential's role may run it. |
| tools[].input_schema | object | JSON Schema of the tool's `args`. |
| invocable_count / totalCount | number | Counts over the full registry. |
Tool input
{ "only_invocable": true, "include_schemas": false }Tool response
{
"tools": [
{ "name": "find_data", "description": "CALL THIS FIRST to locate the real data behind a natural-language concept…", "impact": "read", "capability": "data.read", "can_invoke": true },
{ "name": "query_data", "description": "Run a read-only SQL SELECT over the extracted data…", "impact": "read", "capability": "data.read", "can_invoke": true }
],
"invocable_count": 12,
"totalCount": 48
}Frequently asked questions
Why are some tools listed with can_invoke false?+
An API key runs as a read-only principal, so tools above the `data.read` capability (anything that spends credits or writes) are listed for transparency but cannot be invoked through the API.