Traceway open source release: a full-stack observability platform under the MIT protocol, integrating AI Agent tracing

Core conclusion

Traceway is an open source observability platform native to OpenTelemetry that supports logs, link tracing, metrics, session replay, exception tracking, and AI Agent tracing. It uses the pure MIT protocol (not BSL / Open Core) and supports Docker Compose 90-second self-deployment. Developers simply point the OTLP exporter to it to get started, no Collector or vendor SDK required.

Key Points

• Event time: 2026-05-13 HN Home Page Hot 146 minutes -Affected objects: AI Agent developers, content automation teams, full-stack developers
• Core changes: AI Agent observability moves from "vendor lock-in" to "open source self-hosting"
• GitHub Stars: 615 ★ (on the day of release)

Background and trigger events

On May 13, 2026, the Traceway project appeared on the homepage of Hacker News and received 146 likes. The team aims to be "the only tool you need to know what's going on and how to fix it." Different from most OpenTelemetry solutions on the market (such as Datadog, New Relic, Grafana Cloud), Traceway emphasizes no BSL, no Open Core, no Collector - all functions are included in a binary file.

It is worth mentioning that Traceway’s AI tracking function natively supports mainstream AI providers including OpenRouter, and can track the cost, token usage, latency and complete conversation chain of each LLM call. There is direct value in debugging and optimizing AI Agent workflows.

Key Impact

Adaptation suggestions

For AI Agent developers and content automation team members, Traceway provides three areas of practical value:

1. Debugging AI Agent workflow

• Monitor the token consumption and latency of each LLM call through the AI tracking function
• Utilize link tracing to analyze execution bottlenecks of n8n/OpenClaw workflows
• Use session replay to reproduce user interaction issues during Agent execution

2. Replace commercial monitoring tools

• Datadog’s minimum package is about $200/month, and Traceway self-deployment is completely free
• Directly accept the OpenTelemetry OTLP protocol, and migrate the existing SDK with zero changes
• SHA-256 normalized anomaly aggregation to reduce noise

3. Build the Agent’s observability baseline

• Track the Agent decision chain in the production environment (span waterfall chart) -Set alarm rules to notify when Token consumption or delay exceeds the threshold
• Use the correlation between logs and traces to quickly locate faults

Task List

• Deploy Traceway: docker compose up one-click startup
• Configure OpenTelemetry exporter in n8n Agent
• Set Token consumption alarm threshold
• Establish a review process for Agent execution exceptions

Deployment example

# 90 Traceway
git clone https://github.com/tracewayapp/traceway.git
cd traceway
docker compose up -d

#
curl http://localhost:4318/v1/traces

# : {"partialSuccess":{}}

# OTLP （Python ）
from opentelemetry import trace
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor

provider = TracerProvider()
processor = BatchSpanProcessor(OTLPSpanExporter(endpoint="http://localhost:4318/v1/traces"))
provider.add_span_processor(processor)
trace.set_tracer_provider(provider)

• Traceway GitHub
• Traceway
• OpenTelemetry

OpenAI、n8n、OpenClaw、OpenRouter、Docker、Hermes Agent

Internal link guidance

Want to build an AI automated workflow? See our guide: How to build an automated AI content distribution system with n8n + OpenAI — How to build an AI content automated distribution system with n8n + ChatGPT: a complete 30-minute tutorial

Real case: He used OpenClaw to earn monthly income $14,718 — A complete review of AI helping him start a company — Nat Eliason’s OpenClaw Experiment: $14,718 in 3 Weeks With a $1K AI Budget