Skip to main content
A tool call represents a single tool execution within an agent run. A tool call requires a name before calling .end(). All setter methods return self, so you can chain calls: 
tc.name("get_weather").args({"city": "London"}).result({"temp": 15}).end()

​
Required

​
Create tool call

There are two ways to create a tool call: 
tc = r.tool_call("tool_name")

tc = tcc.tool_call(run_id=r.run_id, tool_name="tool_name")
ParameterRequiredDescription
tool_nameNoThe name of the tool (can be set later with .name())
tool_call_idNoCustom tool call ID. If not provided, a UUID is automatically generated.
import contextcompany as tcc

r = tcc.run()

# Option 1: Create a tool call from a run
tc = r.tool_call("get_weather")

# Option 2: Create a tool call directly with a run ID
tc = tcc.tool_call(run_id=r.run_id, tool_name="get_weather")

​
Set name

The name is required before calling .end(). If you didn’t provide it when creating the tool call, set it with: 
tc.name(tool_name)
ParameterRequiredDescription
tool_nameYesThe name of the tool
tc.name("get_weather")

​
End tool call

End the tool call. A tool call must have a name set before calling .end(). The tool call cannot be modified after calling .end(). 
tc = r.tool_call("get_weather")
tc.args({"city": "London"})

# --- Your tool execution ---
result = get_weather("London")
# --- End of tool execution ---

tc.result(result)
tc.end()

​
Optional tool call data

​
Set arguments

Set the arguments passed to the tool. Values are automatically serialized to JSON if a dictionary is provided. 
tc.args(value)
ParameterRequiredDescription
valueYesArguments passed to the tool (string or dictionary)
# As a dictionary (auto-serialized)
tc.args({"city": "London", "unit": "celsius"})

# As a string
tc.args('{"city": "London", "unit": "celsius"}')

​
Set result

Set the result returned by the tool. Values are automatically serialized to JSON if a dictionary is provided. 
tc.result(value)
ParameterRequiredDescription
valueYesResult returned by the tool (string or dictionary)
# As a dictionary (auto-serialized)
tc.result({"temp": 15, "unit": "C", "condition": "cloudy"})

# As a string
tc.result('{"temp": 15, "unit": "C", "condition": "cloudy"}')

​
Mark tool call as failed

Sets the status code to error. The tool call cannot be modified after calling .error(). Unlike .end(), calling .error() does not require a name to be set. 
tc.error(status_message)
ParameterRequiredDescription
status_messageNoA description of the error
import contextcompany as tcc

r = tcc.run()
r.prompt("What is the weather in London?")

tc = r.tool_call("get_weather")
tc.args({"city": "London"})

try:
    result = get_weather("London")
    tc.result(result)
    tc.end()
except Exception as e:
    tc.error(str(e))

# Run still ends successfully even if a tool call fails
r.end()