eval-analysis
OfficialDiagnose eval signals and improve prompts safely.
Data & Analytics#data-driven#analyze#anti-bias#signal-quality#eval-results#prompt-config#eval-gating
Authoreforge-build
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill analyzes eval results to understand signal quality and guide prompt/config changes while enforcing anti-bias safeguards.
Core Features & Use Cases
- Gather recent eval data with eval_runs to identify run IDs, timestamps, and included scenarios.
- Pull observations via eval_observations for detailed per-scenario signal data and metadata.
- Drill into affected scenarios with eval_scenario_detail to inspect inputs, scoring criteria, and notes, then read relevant prompts and config sources.
- Apply anti-bias gating rules to propose safe, targeted changes and document counter-scenarios.
Quick Start
Analyze the latest eval_run data to surface signal quality and guide prompt/config adjustments.
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: eval-analysis Download link: https://github.com/eforge-build/eforge/archive/main.zip#eval-analysis 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.