Quick Start¶

0. One-Liner Check (Fastest)¶

No setup, no class instantiation:

from longtracer import check

result = check(
    "The Eiffel Tower is 330 meters tall and located in Berlin.",
    ["The Eiffel Tower is a wrought-iron lattice tower in Paris, France. It is 330 metres tall."]
)

print(result.verdict)   # "FAIL"
print(result.summary)   # "1/1 claims supported, 1 hallucination(s) detected."

Or from the command line:

longtracer check "The Eiffel Tower is in Berlin." "The Eiffel Tower is in Paris."
# ✗ FAIL  trust=0.50  hallucinations=1

1. Direct Verification¶

The simplest usage — no setup required:

from longtracer import CitationVerifier

verifier = CitationVerifier()

result = verifier.verify_parallel(
    response="The Eiffel Tower is 330 meters tall and located in Berlin.",
    sources=[
        "The Eiffel Tower is a wrought-iron lattice tower in Paris, France.",
        "It stands 330 metres tall on the Champ de Mars."
    ]
)

print(f"Trust score:   {result.trust_score:.2f}")
print(f"Hallucinations: {result.hallucination_count}")
print(f"All supported:  {result.all_supported}")

for claim in result.claims:
    status = "✓" if claim["supported"] else "✗"
    print(f"  {status} {claim['claim'][:80]}")

2. With Source Metadata¶

Pass metadata to trace which document each claim came from:

result = verifier.verify_parallel(
    response="Paris is the capital of France and has a population of 2 million.",
    sources=["Paris is the capital and most populous city of France."],
    source_metadata=[{"source": "geography.pdf", "page": 12}]
)

for claim in result.claims:
    print(claim["claim"])
    print(f"  Best source: {claim['best_source_metadata']}")
    print(f"  Score: {claim['score']:.3f}")

3. With Tracing Enabled¶

Enable tracing to persist verification results across runs:

from longtracer import LongTracer, CitationVerifier

LongTracer.init(backend="sqlite", verbose=True)
tracer = LongTracer.get_tracer()

verifier = CitationVerifier(tracer=tracer)

with tracer.span("my_pipeline"):
    result = verifier.verify_parallel(
        response="...",
        sources=["..."]
    )

# View traces later
# longtracer view --last

4. Multi-Project Tracing¶

Track multiple RAG applications independently:

from longtracer import LongTracer

LongTracer.init(project_name="chatbot-prod", backend="sqlite")
LongTracer.init(project_name="search-api", backend="sqlite")

chatbot = LongTracer.get_tracer("chatbot-prod")
search  = LongTracer.get_tracer("search-api")

# Each project's traces are stored and filterable separately

5. View Traces via CLI¶

longtracer view                        # list recent traces
longtracer view --last                 # view most recent
longtracer view --project chatbot-prod # filter by project
longtracer view --html <trace_id>      # export HTML report