Claudius
Status

modelharness

Plugin

Claude Code plugin providing a zero-configuration behavioral harness with a reproducible 408-run benchmark for model cost and quality optimization.

by vitaliikapliuk·vitaliikapliuk/modelharness·Python·v0.3.0
76· B
Install
/plugin marketplace add vitaliikapliuk/modelharness
Stars
51
7d change
1.92%
Downloads / week
Last active
1mo ago
About

Make every model cheaper or better. Measured on four Claude models — none got worse.

What it actually is: a zero-config Claude Code plugin. On every session start it injects a ≈910-token behavioral core — six working practices distilled from how Fable 5 was trained to operate — plus three on-demand skills and a fresh-context verifier agent. No commands to learn; the model just starts working differently:

Read more on GitHub →
What's inside
3 commands/goal/retro/verify
1 agentverifier
3 skillsdelegation-triggersmemory-disciplineverification-loop
30-day stars
Trust factors
Source
community
Known advisories
0
Maintenance
maintained
License
MIT
Age
1 months
developer-toolspromptingtoken-efficiency#benchmark#model-optimization#behavioral-harness#cost-reduction#haiku#sonnet#opus#agent-harness#claude-code-plugin#fable#llm-harness