llm-monitoring-dashboard
Auto-generates an LLM usage monitoring page in a PM admin dashboard. Tokuin CLI-based token/cost/latency tracking + user ranking system + inactive user tracking + data-driven PM insights + Cmd+K global search + per-user drilldown navigation. Supports OpenAI/Anthropic/Gemini/OpenRouter.
Install with Tessl CLI
npx tessl i github:supercent-io/skills-template --skill llm-monitoring-dashboard67
Quality
52%
Does it follow best practices?
Impact
99%
1.86xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agent-skills/llm-monitoring-dashboard/SKILL.mdEvaluation results
97%
55%LLM Usage Metrics Collection Pipeline
Privacy-safe metrics collection pipeline
No hardcoded API keys
100%
100%
dry-run default
0%
100%
JSONL output path
100%
100%
Prompt hashing
0%
100%
Prompt categorization
100%
100%
Keyword-based classification
100%
100%
User context fields
0%
100%
Latency fields
0%
100%
is_dry_run field
0%
62%
.env in .gitignore
100%
100%
Tokuin CLI usage
0%
100%
status_code field
0%
100%
Without context: $0.3721 · 1m 43s · 18 turns · 25 in / 6,437 out tokens
With context: $0.8307 · 2m 33s · 35 turns · 329 in / 9,406 out tokens
100%
43%LLM Usage Monitoring Dashboard
Dashboard design tokens and drilldown navigation
Background token --bg-base
50%
100%
CSS design token variables
100%
100%
3-level traffic light system
100%
100%
Traffic light hex values
0%
100%
Badge elements for status
100%
100%
Monospace font for metrics
0%
100%
Tabular numeric alignment
0%
100%
User drilldown navigation
100%
100%
Auto-generated PM insights
100%
100%
Chart.js for charts
100%
100%
Colorblind-friendly series colors
0%
100%
Ranking medal colors
0%
100%
Without context: $0.6550 · 3m 13s · 14 turns · 21 in / 16,414 out tokens
With context: $0.9212 · 3m 5s · 21 turns · 281 in / 13,966 out tokens
100%
39%LLM Cost Monitoring: Automated Reporting and Alerting
PM report automation and cost alerting
safety-guard.sh exists
87%
100%
PM report Markdown format
100%
100%
User ranking by cost
100%
100%
Inactive user section
100%
100%
Adoption rate metric
50%
100%
Report filename pattern
25%
100%
95% SLA success rate check
12%
100%
Configurable cost threshold
70%
100%
Alert exit code 1
100%
100%
Slack webhook from env var
100%
100%
Weekly report cron Monday 9am
0%
100%
Cost alert cron hourly
0%
100%
Without context: $0.5762 · 2m 50s · 22 turns · 29 in / 11,101 out tokens
With context: $1.8245 · 5m 37s · 48 turns · 54 in / 22,744 out tokens
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.