Examples

Basic Capture

from briefcase import capture

@capture()
def summarize(text: str) -> str:
    return llm.complete(f"Summarize: {text}")

result = summarize("Long document text...")

Capture with Storage

from briefcase import capture, configure
from briefcase.storage import SQLiteBackend

configure(storage_backend=SQLiteBackend("./decisions.db"))

@capture()
def classify(text: str) -> str:
    return llm.classify(text)

Replay and Compare

from briefcase.replay import ReplayEngine
from briefcase.drift import DriftDetector

engine = ReplayEngine()
detector = DriftDetector()

snapshot = engine.load("snapshot-id")
replayed = engine.replay(snapshot)

metrics = detector.compare(snapshot, replayed)
print(f"Similarity: {metrics.similarity_score}")

Multi-Agent Workflow

from briefcase.correlation import CorrelationContext

with CorrelationContext(workflow_id="pipeline-001") as ctx:
    research = researcher_agent.run("Find recent papers on X")
    analysis = analyst_agent.run(research)
    report = writer_agent.run(analysis)

More Examples

See the examples directory for complete working examples including validation, lakeFS versioning, and multi-agent correlation.