Telemetry shows what's working and what isn't, so you have the data you need to make the right decisions. The latest Zenable CLI captures coding-agent tool-use telemetry on every hook fire -- which tools the agent reached for, what it tried to write, when it retried, when it gave up, when an action looks anomalous -- and ships it to a centralized observability view per tenant.
Use the data to decide whether a particular agent or repo needs sandboxing, whether a guardrail is being tripped more than expected, or whether the model an agent is routing to is the right fit for the workload. The signal is per-IDE, so you can compare patterns across Claude Code, Cursor, Windsurf, Kiro, VS Code Copilot, and Auggie without asking developers to change anything about how they work.
We're incrementally rolling this out. If you want early access for your tenant, drop us a note at hello@zenable.io and we'll flip it on.