Framework	JS Import	Function
LangChain	`risicare/langchain`	`RisicareCallbackHandler` + `withSuppression()`
LangGraph	`risicare/langgraph`	`instrumentLangGraph()`
Instructor	`risicare/instructor`	`patchInstructor()`
LlamaIndex	`risicare/llamaindex`	`RisicareLlamaIndexHandler`

What's Captured

Framework integrations provide rich observability beyond basic LLM tracing:

Feature	Description
Agent Identity	Name, role, and hierarchy
Decision Flow	Think/Decide/Act phases
Tool Execution	Tool calls with inputs/outputs
State Changes	Graph state transitions
Inter-Agent Messages	Communication between agents
Iteration Tracking	Loop counts and convergence

Provider Span Behavior

Some frameworks automatically suppress underlying LLM provider spans to avoid duplicate traces. Others let provider spans appear alongside framework spans for full visibility.

Framework	Provider Suppression	Notes
LangChain	Yes	Callback dedup prevents double LLM spans
LiteLLM	Yes	Context manager suppresses provider instrumentation
DSPy	Yes	Suppresses for LM calls
LlamaIndex	Selective	Suppresses LLM and embedding spans only
LangGraph	No	LangChain callbacks handle dedup when both active
CrewAI	No	Provider spans appear as children
AutoGen	No	Provider spans appear as children
OpenAI Agents	No	Provider spans appear as children
Instructor	No	Provider spans appear as children
Pydantic AI	No	Provider spans appear as children

Auto vs Manual Instrumentation

Auto-Instrumentation

Enable with a single import:

import risicare
risicare.init()
 
# Framework code is automatically traced
from langgraph.graph import StateGraph
# ... your agent code

Manual Instrumentation

Add explicit context for more control:

from risicare import agent, trace_think, trace_decide, trace_act
 
@agent(name="researcher", role="specialist")
def research_agent(query: str):
    @trace_think
    def analyze():
        return analyze_query(query)
 
    @trace_decide
    def plan():
        return create_plan(analysis)
 
    @trace_act
    def execute():
        return run_search(plan)
 
    analysis = analyze()
    plan = plan()
    return execute()

Framework Detection

Risicare automatically detects which frameworks are installed and instruments them:

import risicare
 
# Returns set of instrumented module names
print(risicare.get_instrumented_modules())
# ['langgraph', 'crewai', 'autogen']

Next Steps

LangGraph Integration

Detailed LangGraph guide

Learn more

Multi-Agent Observability

Understand agent interactions

Learn more

Edit this page on GitHub

PreviousVercel AI (JS)NextLangGraph