---
schema_version: "newruntime-topic-hub-v0.2"
type: "topic_hub"
stable_id: "topic_hub:agent-observability"
slug: "agent-observability"
title: "Agent Observability - New Runtime"
description: "A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent observability."
retrieval_nugget: "A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent observability. Agent Observability is tracked here as an evidence-linked topic, not as a static glossary entry. The page connects raw observations to pattern hypotheses, longer analysis, and public sources. Use it as the canonical landing page before drilling into individual records."
answer: ["Agent Observability is tracked here as an evidence-linked topic, not as a static glossary entry.","The page connects raw observations to pattern hypotheses, longer analysis, and public sources.","Use it as the canonical landing page before drilling into individual records."]
search_intents: ["agent observability","agent observability AI agents","agent observability software"]
status: "tracked"
last_updated: "2026-08-03"
record_date: "2026-08-03"
date_kind: "last_updated"
counts: {"total":6,"signals":0,"patterns":0,"posts":6,"atlas":0,"sources":6}
routes: {"html":"https://newruntime.com/topics/agent-observability/","markdown":"https://newruntime.com/topics/agent-observability.md","json":"https://newruntime.com/topics/agent-observability.json"}
source_urls: ["https://aws.amazon.com/blogs/machine-learning/optimizing-production-agents-with-amazon-bedrock-agentcore-observability","https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga","https://github.com/jarrodwatts/claude-hud","https://jamwithai.substack.com/p/build-your-own-job-agent-part-1","https://www.comet.com/site/blog/debugging-ai-agents","https://www.comet.com/site/blog/digibee-opik-user-story"]
top_sources: ["https://aws.amazon.com/blogs/machine-learning/optimizing-production-agents-with-amazon-bedrock-agentcore-observability","https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga","https://github.com/jarrodwatts/claude-hud","https://jamwithai.substack.com/p/build-your-own-job-agent-part-1","https://www.comet.com/site/blog/debugging-ai-agents","https://www.comet.com/site/blog/digibee-opik-user-story"]
---

# Agent Observability - New Runtime

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent observability.

## Retrieval answer

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent observability. Agent Observability is tracked here as an evidence-linked topic, not as a static glossary entry. The page connects raw observations to pattern hypotheses, longer analysis, and public sources. Use it as the canonical landing page before drilling into individual records.

## Patterns


## Field Notes

- [Opik Turns Agent Traces Into a Debugging Surface](https://newruntime.com/posts/opik-agent-diagnostics-trace-layer/): Opik's agent diagnostics frame tracing as an operational debugging layer, not just a transcript viewer for individual runs.
- [Prompt Versioning Is Becoming Agent Operations](https://newruntime.com/posts/digibee-opik-prompt-versioning-loop/): The Digibee and Opik case shows prompt management moving into the same traceable release loop as code, evals, and production incidents.
- [The Observable Job Agent Is a Useful Agent-Product Template](https://newruntime.com/posts/observable-job-agent-opik-langgraph/): The Observable Job Agent combines LangGraph state, model-chosen tools, bounded search loops, and Opik traces into a small agent product pattern.
- [AgentCore Makes Slow Agents A Traceable Operations Problem](https://newruntime.com/posts/aws-agentcore-observability-performance-budgets/): AWS shows how AgentCore Observability and CloudWatch traces expose latency, sequential tools, memory growth, and context accumulation in production agents.
- [Google Runs The Same Agent Metrics Before And After Launch](https://newruntime.com/posts/google-agent-evaluations-offline-online/): Gemini Enterprise Agent Platform makes experiments, adaptive rubrics, trace review, simulations, online monitors, and drift alerts generally available on one evaluation engine.
- [Claude HUD Turns Agent Observability into a Status Bar](https://newruntime.com/posts/claude-hud-agent-observability/): Claude HUD shows that agent observability can begin as a small status line for context health, tools, running agents, and progress instead of a large dashboard.

## Recent Raw Signals


## Public Sources

- https://aws.amazon.com/blogs/machine-learning/optimizing-production-agents-with-amazon-bedrock-agentcore-observability
- https://developers.googleblog.com/agent-and-model-evaluations-in-gemini-enterprise-agent-platform-are-now-ga
- https://github.com/jarrodwatts/claude-hud
- https://jamwithai.substack.com/p/build-your-own-job-agent-part-1
- https://www.comet.com/site/blog/debugging-ai-agents
- https://www.comet.com/site/blog/digibee-opik-user-story