Clone to Claude
Last commit on May 3, 2026·Created on March 29, 2026
Lomnus-ai/TokenBurner
“Load-tests an LLM backend by burning thinking tokens on demand.”
Combined rank
#436
across all skills
In Testing
#4
category rank
Stars
179
0.0% in last 7d
Forks
14
0.0% in last 7d
Watchers
17
0.0% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
Description
This skill makes Claude solve dozens of complex math problems in extended thinking before every response, turning every conversation into a controlled load test. It helps AI backend engineers, research teams, and platform engineers measure latency, validate autoscaling triggers, model cost under peak usage, and A/B test backend configuration changes with predictable token consumption.
WHO IT'S FOR
LLM Ops / Reliability Engineers
stress test LLM backend infrastructure
AI Agent / Automation Builders
simulate high token consumption scenarios
Platform / DevEx Teams
evaluate LLM cost and performance boundaries
Backend / Infrastructure Engineers
validate LLM response consistency under load
30 Days of GitHub Stars
+2 stars↓ 2.2% monthly change
Jul 7Jul 14Jul 21Jul 28Aug 5
Compatibility & Install1 Skill · 1 Compatible agent · 3 categories · 2 install paths
Compatible agents
❯Claude Code
Categories
TestingDeveloper ToolingToken Optimization
Also in this category4 of 4