featuresMay 5, 2026· 3 min read

CINFO and the dashboard: a cache you can actually watch work

Infrastructure you can't observe is infrastructure you don't trust. CINFO and the built-in dashboard expose hit rate, saved spend, safety blocks, memory pressure, and license state in real time.

The first question every operator asks a new piece of infrastructure is 'what is it actually doing right now?', and the worst answer is 'trust us.' Crowkis answers with surfaces instead: CINFO returns a Crowkis-flavoured INFO with server, cache, savings, security, db, and license sections, and the built-in dashboard renders the same truth live.

In plain words: Crowkis ships with a live dashboard and an INFO command showing hit rate, money saved, safety blocks, and memory use, plus Prometheus metrics, so you can watch it work instead of taking it on faith.

What you watch is the stuff that matters operationally: the live hit rate and the breakdown of hit types, the dollars saved against the dollars spent, the safety blocks by pipeline stage, memory and block-cache pressure against their limits, and, because it gates features, the license plan, fingerprint, and days remaining. Agent-memory and AI-gateway metrics join the same view.

the crowkis read path, five gates, every one can veto

1
incoming query
2
intent classifier
3
template match
4
HNSW neighbours
5
confidence gate
6
trust + freshness
7
answer · <1ms
8
(nil) → your model

Reuse only when meaning, structure, confidence, and trust all agree.

It's all exposed three ways for the three audiences: the embedded dashboard for a human glancing at it, /api/metrics as JSON for your own tooling, and /metrics in Prometheus format for the monitoring stack you already run. The same numbers, shaped for whoever's asking, no separate agent to install, because it ships in the binary.

The bottom line

Observability that comes free in the box, rather than as a bolt-on you assemble, is what lets a cache earn the critical path. You don't trust Crowkis because the README says to; you trust it because you can watch it save you money and block the bad writes, in real time, on a dashboard that was already running.