anthropic.callable_tools()
Give Anthropic’s tool runner access to an existing Console session.
Usage
anthropic.callable_tools(console, **kwargs)The runner calls Console’s tools and returns their output to Claude before requesting the next response. Keep the connection open for the whole conversation so later tool calls can use variables from earlier cells. A synchronous client produces synchronous tools; an async client produces async tools for AsyncAnthropic.
Text results return text, and text-only tool errors raise RuntimeError. Results containing images use Anthropic content blocks with the server’s image data and MIME types, preserving the order of text and images. An image result marked as an error raises anthropic.lib.tools.ToolError carrying those blocks, so the tool runner retains the images and error status.
Discovery includes every tool, even when the server has no send tool.
Parameters
console: MCPConsole | AsyncMCPConsole-
A connected
MCPConsoleorAsyncMCPConsole. **kwargs: Any-
Options passed to Anthropic’s
beta_toolorbeta_async_tool, such asdefer_loading.
Returns
list[Any]-
All discovered tools for
client.beta.messages.tool_runner(tools=...).
Examples
Install with pip install 'mcp-console[anthropic]'. Set ANTHROPIC_API_KEY and set ANTHROPIC_MODEL to the Claude model ID you want to use. This script prints the model’s final answer:
import os
import mcp_console
from anthropic import Anthropic
from mcp_console import MCPConsole
with Anthropic() as client, MCPConsole() as console:
runner = client.beta.messages.tool_runner(
model=os.environ["ANTHROPIC_MODEL"],
max_tokens=4096,
tools=mcp_console.anthropic.callable_tools(console),
messages=[
{
"role": "user",
"content": "Use Python in Console to calculate the mean "
"and sample standard deviation of [12, 15, 18, 20, 25].",
}
],
)
message = runner.until_done()
for block in message.content:
if block.type == "text":
print(block.text)