Agent observability tooling is lacking: devs want a local OTEL service for traces
Cautious_Chicken_604 · reddit · 2026-09-08
A developer argues agent observability tooling is severely lacking for mining LLM calls, tool calls, and file I/O to engineer out failure modes. dsh's built-in trajectory tab is closest to what they want, but they seek a locally-hosted standalone service with OTEL export, custom spans, and a session-based UI showing agent/subagent file reads, writes, and tool calls. They tried and dropped Opik (no opencode support, poor UX), considered cobbling together their Cstack of Seq/Jaeger/Grafana, and ruled out vibecoding a custom solution. The post doubles as a community survey on what others use.
More from coding & agent
- AI Operates Lab Robot: Claude Automates Yeast Colony Picking — nlarusstone · 2026-09-08
- Open-Source Chalkboarding Skill: AI Creates Chalkboard-Style Teaching Animations — Madisonkanna · 2026-09-08
- What evidence is strong enough to rule out a step when debugging agent workflows? — Sensitive-Parsnip-12 · 2026-09-08
- Notch demos AI-generated sparse voxel octree renderer without describing a single render pass — anselm · 2026-09-08
- Production teams confront the missing piece for autonomous agents: a reversibility layer to roll back mistakes — Anxious-Variation508 · 2026-09-08
- "I have not met anyone above 100IQ who runs a gorillion agents constantly," says AI commentator — examachine · 2026-09-08