kenny-telemetrylisted

# kenny telemetry Telemetry is a **pushed** snapshot: the agent collects sections on a timer and sends one `telemetry` frame; the server stores it and evaluates fleet health. On-demand refresh is the `telemetry_collect` tool. See `docs/adr/0007-telemetry-push-model-and-sqlite-storage.md`. ## The pieces - **Agent collector** (`kenny-agent/src/telemetry/collectors/<section>.rs`): produces one section payload — required `status` + `summary`, plus raw fields defined in the contract. Gate Windows-only data behind `#[cfg(windows)]`; provide a portable fallback (often `sysinfo`) or a `status:"ok", summary:"n/a on this platform"` stub. - **Scheduler** (`kenny-agent/src/telemetry/scheduler.rs`): runs all collectors every interval (default 900 s) and pushes the snapshot. - **Store** (`kenny-server/kenny_server/store.py`): SQLite, latest snapshot + ~30 days of history per agent, with a retention job. - **Health rules** (`kenny-server/kenny_server/health_rules.py`): the authoritative thresholds (e.g. disk > 90% ⇒ crit, Defender scan > 14 days ⇒ warn). Aggregates section statuses into an overall per-agent health. - **Dashboard** (`kenny-server/kenny_server/webui/`): fleet grid of agent tiles (overall traffic-light), drill-down to one agent's sections + history trends, "refresh now". ## Sections Mandatory + four bundles (hardware health, security & crypto, update & stability, operations & daily) — the authoritative list is in `docs/protocol.md` § Telemetry sections. Us