
Your Agents Trust Your Telemetry. Should They?
OllyGarden and Causely on the telemetry that fools agents: what we found in real data, and why an agent can read accurate signals and still blame the wrong service.
- AI
- Observability
- Instrumentation
Guides, customer stories, articles, and talks from the team behind OllyGarden, written for engineers who run OpenTelemetry® in production. Start with what telemetry quality means, see how it is measured, and read what changed for real teams.
Telemetry quality is how well your OpenTelemetry data helps you run your systems. Learn its five dimensions, the signs of bad telemetry, and how to measure it.
Read the guideThe Instrumentation Score rates OpenTelemetry instrumentation from 0 to 100. See how it is calculated, all 25 rules, and a worked example.
See the scoreTell us your observability backend, bill and stack. See the telemetry problems teams like yours usually have and what they may cost you.
Start the quick checkOpen source OpenTelemetry agent skills for Claude Code, Cursor, and Codex: 18 vendor-neutral fact skills and 5 OllyGarden skills, free under Apache 2.0.
Install the skillsAnonymized but real OllyGarden customer stories: what bad telemetry we found, what changed at the source, and the numbers that moved.
Read the storiesPractical guides on OpenTelemetry instrumentation, Collector configuration, naming conventions, and telemetry cost from the OllyGarden team.
Browse the blogNews and announcements from OllyGarden, the company fixing bad OpenTelemetry telemetry, including product launches and partnerships.
Read the newsWebinars, demos, and conference talks on OpenTelemetry, telemetry quality, and instrumentation from the OllyGarden team.
Watch the talksWhere to meet the OllyGarden team next: OpenTelemetry and observability conferences, meetups, and community days.
See upcoming events10 questions, one per OpenTelemetry topic, that challenge even the project's maintainers.
Take the quiz10 real, anonymized OpenTelemetry cases and the conclusions drawn from them. Which hold up?
Play the gameAnswers about OllyGarden products, data privacy, rate limits, and pricing.
Read the FAQLogos, colors, Space Grotesk, icons, and photography for the OllyGarden brand, with the rules for using each and files ready to download.
Open the kitProduct guides for Insights and Rose, and the OllyGarden API reference.
Open the docs
OllyGarden and Causely on the telemetry that fools agents: what we found in real data, and why an agent can read accurate signals and still blame the wrong service.

The smallest working continuous profiling pipeline we know of: the OpenTelemetry eBPF profiler on a local kind cluster, sending flame graphs to Grafana's otel-lgtm stack, plus the one setting kind needs that a cloud cluster does not.

Coding agents write the code and AI SRE agents operate it, so telemetry is the source of truth between them. Rose now brings low-volume, high-quality OpenTelemetry instrumentation to every repository.