skim
Token Intelligence Dashboard — sign in to continue
Invalid credentials. Try again.
or continue with
skim
Token Intelligence
Overview
Dashboard Sessions
Analysis
Usage Models Savings
Admin
Team Insights Settings
U
Loading…
—
Dashboard
Total Tokens Sent
—
— API calls
Estimated Cost
—
avg per call
Tokens Saved by skim
—
— reduction
Avg Context Fill
—
of model limit
Token Usage Over Time
Usage by Model
Tokens Saved vs Sent
Top Users
Recent API Calls
—
TimeModelTokens In Tokens OutSavedCostStatus
Team Usage Leaderboard
—
#UserTeamCalls Tokens InCostWaste %Cache Hit %Avg Latency
Org Optimization Insights
AI-generated recommendations based on your team's usage patterns.
No insights yet — connect team proxies to start collecting data.
All API Calls
—
#Timestamp (UTC)Model Tokens InTokens OutSavedCached CostLatency
Hourly Activity
Daily Calls
Daily Breakdown
DateCallsTokens In Tokens SavedCostSave %
Tokens by Model
Cost by Model
Model Comparison
ModelCallsTokens InTokens Out CostCost/1k InCache Hit %Waste %Avg Latency
Tokens Saved by skim
—
Estimated $ Saved
—
Saved via Cache Hits
—
Cumulative Savings Over Time
Daily Save Rate (%)
Sent vs Saved vs Cached
Savings by Model
Account Settings

Generate an API key to connect the skim proxy to this dashboard.

Copy this key now — it won't be shown again.
Then set these env vars before running skim proxy: 
export SKIM_SERVER_URL=
export SKIM_SERVER_TOKEN=