trace-replay

Trace Replay and Network Simulation Framework

Local-first, IM-native AI coworker runtime with async role agents, trace replay, evidence, and agentic evaluation.

🔬This program benchmarks the I/O performance per container. You can see the implementation of this program on our website.

Generate reproducible deep-learning workload traces by mapping production GPU-cluster traces to profiled executable workloads.

Real-world evaluation framework for AI coding agents — measures safety, containment, cost, and autonomy beyond just correctness.

Deterministic intermediate representation for AI agents — compile, verify, execute, and replay structured intent.

Record, replay, evaluate, and regression-test AI agents with deterministic traces, calibrated LLM-as-a-judge evals, OpenTelemetry observability, and CI quality gates.

morph-replay-runner is a command-line interface (CLI) tool designed to execute TRACE-REPLAY-KIT bundles with branch-N parallelism on Morph Cloud. This tool streamlines the process of running replay tasks, ensuring efficient and scalable execution with comprehensive evidence collection and CERT-V1 compliance.

Check AI agent projects before they go live: local reports, trace replay, and failure-to-eval for MCP/RAG/LLM apps.

Agent engineering studio with deterministic tool orchestration, policy guardrails, trace replay, and scenario evaluations.

A reference implementation for tracing, replaying, evaluating, and governing agent runs across trust boundaries.

Neutral, reproducible comparison of Go in-process cache libraries: hit rate and per-op cost across synthetic workloads and real traces, with the provenance to regenerate every number.

Governable, reviewable and replayable multi-agent ResearchOps reference module for GOAI 2026 Agent Infrastructure M1.

Deterministic message-delivery state machine and trace auditor for MoonBit

Turn failed agent traces into replayable regression cases.