AI Ops Dashboard
Where it all comes together
Show or hide any dashboard tab.
How: Toggle tabs on/off and pick the default tab. Hidden tabs stay installed and can be turned back on anytime — nothing is deleted.
Change the dashboard's color theme.
How: Pick any of the 8 themes — your choice is saved in this browser only (it doesn't change the shared default).
Sub-tasks
1 / 3
33% shipped
Sprints
2
tracked
Commits · 90d
154
6 active days
Commits · 7d
1
Across all connected repos
Repos active · 7d
1
Repos with ≥1 commit this week
Token Savings
How much context (tokens and dollars) claude-mem + Graphify save versus loading everything every time.
How: Read-only — it reads data/savings.json, refreshed by `npm run data:collect-savings` and a SessionStart hook.
Context saved (claude-mem, cumulative)
$5,661
Directional, not an invoice — these compare against a “load everything, every time” baseline you would never actually run. A sense of scale.
claude-mem
Saved
$5,661
Work banked
189.3M tok
Observations
11,364
Efficiency
99.7%
Work banked over time
Graphify reduction over time
By project
Graphify
Reduction
16.9×
$3.16 saved per question
Naïve corpus vs. graph query
At ~8 graph queries/day, that’s ≈ $759/mo saved.
Graphify keeps no per-query log, so the query rate is an assumption from graphifyAssumedQueriesPerDay — not a measurement.
Build investment ledger
| Graph build tokens | 1M tok |
| Build cost | $31.23 |
| Nodes / edges | 1,682 / 3,088 |
| Build runs | 4 |
The honest investment side — what it cost to build the graph. Each question afterward draws it down by $3.16.
Method: claude-mem saved = work tokens − recall tokens (50 tok/observation). Graphify saved/query = naïve corpus read − average graph-query cost. Dollars = tokens × rate ($30/M tokens, configurable in .ai-ops-dashboard/savings.config.json). Anything marked illustrative is an estimate, not a measurement.