Signals overview

Signal	Where	On by default	Scope
Metrics (Prometheus text 0.0.4)	`serve --metrics-addr` and UI `GET /metrics`	In memory, always; exposed only when asked	The process that serves them
Logs (`log/slog`, text or JSON)	stderr	Yes, at `info`	The process
Call log and query log	`call_log` and `query_log` tables	No (`MEMO_QUERY_LOG=1`)	The file: every process writing to it
Snapshot (`memo-mcp metrics`)	stdout	On demand	The file: gauges plus log statistics
Audit	`audit` table	Always	The file: every write

The per-process caveat

The MCP server, the UI and each CLI command are separate processes. Counters such as memo_mcp_tool_calls_total live in the memory of the process that served the calls:

A server process exists only while its client session is open. Its counters start at zero with each session. Scrape it while it runs; Prometheus rate() handles the resets.
The UI has its own counters, mostly memo_ui_*. Like every process, it also serves the memo_kb_* gauges, which are read from the file.
History across sessions comes from the file: memo-mcp metrics (with MEMO_QUERY_LOG=1) and the memo_kb_* gauges.

Recommended setups

Goal	Setup
Glance at health now	`memo-mcp metrics`, or `memo-mcp status`
Trends of knowledge-base size, backlog and pages	Run `memo-mcp ui --addr 127.0.0.1:9470` as a long-lived process and scrape its `/metrics` for `memo_kb_*`
Live tool and search latency, errors, degraded mode	Add `--metrics-addr` to one server entry and scrape it while sessions run
Per-call history and slow-call forensics	`MEMO_QUERY_LOG=1`, then `memo-mcp log calls` and `memo-mcp metrics --since 7d`

Naming

Prefix memo_; counters end in _total; units are suffixes (_seconds, _bytes), and seconds are never milliseconds.
Labels are bounded enums: tool names, outcomes, arms, models, namespaces. Never query text, ids or paths.
OpenTelemetry mapping: replace _ with . and drop _total. For example, memo_mcp_tool_calls_total maps to memo.mcp.tool_calls.