evolve

Evolve top-ranked hypotheses into testable versions with provenance and Elo scores.

Updated Aug 27, 2026

Search Tags:#automation #research #provenance #hypothesis #knowledge-graph #elo #evolve

Ready to run

npx skills add https://github.com/achousal/EngramR --skill evolve-achousal

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: evolve
Source: https://github.com/achousal/EngramR/tree/main/.claude/skills/evolve
Command: npx skills add https://github.com/achousal/EngramR --skill evolve-achousal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Refine and evolve top-ranked hypotheses into stronger, more actionable versions to accelerate scientific progress.

Core Features & Use Cases

Evolution modes: grounding enhancement, combination, simplification, research extension, and divergent exploration to adapt hypotheses to new evidence.
Auto-generation scaffolds: creates generation notes with parent links, Elo scores, and clear evolution history for traceability.
Governance and provenance: links between parent and child notes, and incorporation of meta-reviews into the evolution prompts.

Quick Start

Provide the parent hypothesis IDs and choose an evolution mode to generate and review an evolved hypothesis.

Frequently Asked Questions about evolve

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I refine top-ranked hypotheses into stronger, testable versions?▼

To refine top-ranked hypotheses, you provide parent hypothesis IDs and select an evolution mode to generate stronger, testable versions with documented lineage and provenance.

What evolution modes are available for adapting hypotheses to new evidence?▼

Available evolution modes for adapting hypotheses include grounding enhancement, combination, simplification, research extension, and divergent exploration.

Can I track the provenance and Elo scores of evolved hypotheses automatically?▼

Yes, the system auto-generates scaffolds that create notes with parent links, Elo scores, and clear evolution history to trace the provenance of evolved hypotheses.

How do meta-reviews factor into the hypothesis evolution workflow?▼

Meta-reviews are incorporated directly into the evolution prompts to guide the refinement process and ensure new hypotheses align with existing knowledge-graph-backed evidence.

What is the best way to combine multiple ranked hypotheses into a single stronger note?▼

The best way to combine multiple ranked hypotheses is using the combination evolution mode, which merges parent hypotheses while enforcing generation lineage and documented reasoning.

Does hypothesis evolution work without a pre-existing knowledge graph?▼

No, this evolution workflow applies to a knowledge-graph-backed hypothesis system that uses meta-reviews and Elo rankings to guide refinement and provenance tracking.