Agents

Track individual agent performance and inter-agent interactions.

Agent Identity

Define agent identity in your code:

from risicare import agent
 
@agent(name="researcher", role="specialist")
def research_agent(query: str):
    # Agent code here
    pass

Or with context managers:

from risicare import agent_context
 
with agent_context(
    "planner-001",                   # agent_id (required, positional)
    agent_name="planner",
    agent_role="orchestrator",
    agent_type="custom",
    version=1,
):
    # Agent code here
    pass

Agent Attributes

Agent Roles

Agent Dashboard

The KPI strip shows: Total Agents, Invocations (with trace count), Success Rate (with errors), Avg Latency (with P95), Total Errors (with error rate), and Total Cost (with tokens).

Agent List

View all agents with metrics:

Agent Detail

Deep dive into a single agent:

The agent page shows KPI cards and three tabs: Recent Traces, Tool Usage and Error Breakdown.

Agent Hierarchy

Track parent-child relationships:

@agent(name="orchestrator", role="orchestrator")
def orchestrator():
    with agent_context(
        "researcher-001",
        agent_name="researcher",
        agent_role="specialist",
    ):
        research()
 
    with agent_context(
        "writer-001",
        agent_name="writer",
        agent_role="specialist",
    ):
        write()

Visualized as:

orchestrator (orchestrator)
├── researcher (specialist)
└── writer (specialist)

Inter-Agent Communication

Track messages between agents:

from risicare import trace_message
 
@trace_message(target="researcher-001", target_name="Researcher")
def send_research_request(task: str):
    return researcher.process(task)

Message Types

Message Attributes

The published Python SDK records the target agent on message and delegation spans:

The SDK does not record sender, receiver, message_type, content or correlation_id attributes; set them yourself with set_attribute() if you need them.

Agent Delegation

Track when agents delegate tasks:

from risicare import trace_delegate
 
@trace_delegate(target="executor-001", target_name="Executor")
def delegate_execution(plan: dict):
    return executor.execute(plan)

Agent Analytics

Topology

The Topology tab of the Agents page shows the agents as a sunburst or as a treemap, grouped by name prefix. The dashboard has no network graph of agent interactions.

Agent Iterations

Record iteration counts on iterative agent loops via agent metadata. (Note: this stores the value in the span's metadata; the dedicated iteration / max_iterations agent analytics are not yet auto-populated — see the in-flight note above.)

@agent(name="refiner", role="specialist")
def iterative_refine(content: str, max_iterations: int = 3):
    for i in range(max_iterations):
        with agent_context(
            "refiner-001",
            agent_name="refiner",
            metadata={"iteration": i + 1}
        ):
            content = refine_step(content)
            if is_good_enough(content):
                break
    return content

Filtering Agents

The Management API is not available with an API key during the beta, so you cannot filter agents over HTTP. There is no field:value query language — a filter such as name:researcher or error_rate:>0.1 is not supported.

Next Steps