harness-improvement-loop
CommunityAutomate harness improvements from runtime signals.
Authortylerjrbuell
Version1.0.0
Installs0
System Documentation
What problem does it solve?
The Harness Improvement Loop module enables teams to continuously refine a live AI harness by running instrumented probes, collecting runtime outputs, and surfacing concrete, templated improvements for the harness.
Core Features & Use Cases
- Live probe orchestration: runs five baseline probes against the kernel and strategies to reveal weaknesses in loop handling, context pressure, and tool usage.
- Structured reporting: produces a detailed improvement report template that aggregates observations across passes.
- Evolution-ready: integrates with agent-tdd to implement fixes while preserving historical data and baselines.
- Observability-driven optimization: identifies coverage gaps, regressions, and measurable success criteria for next-pass focus.
Quick Start
Start a new harness-improvement pass by copying the REPORT-TEMPLATE.md to harness-reports and running the probe suite.
Dependency Matrix
Required Modules
@reactive-agents/runtimeollama@reactive-agents/llm-provider@reactive-agents/toolseffect
Components
scripts
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: harness-improvement-loop Download link: https://github.com/tylerjrbuell/reactive-agents-ts/archive/main.zip#harness-improvement-loop Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.