benchmark-driven-improvement
OfficialSystematic benchmark fixes for Serf.
Authorprime-radiant-inc
Version1.0.0
Installs0
System Documentation
What problem does it solve?
Diagnoses and iterates on Serf benchmark failures to improve autonomous agent reliability.
Core Features & Use Cases
- Provides a structured workflow for extracting, reproducing, and fixing benchmark failures.
- Facilitates transcript analysis, session interrogation, and iterative code/prompt improvements to generalize fixes across tasks.
- Supports local execution, tool usage assessment, and verification through standard evaluation tooling.
Quick Start
Run a local benchmark investigation workflow against a cached task, then review transcripts and iterate on prompts, tools, and code to improve robustness.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 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: benchmark-driven-improvement Download link: https://github.com/prime-radiant-inc/serf/archive/main.zip#benchmark-driven-improvement 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 223,000+ vetted skills library on demand.