Back to Skills

Corpus2Skill

SkillGitHubActivePython

Official Findings of EMNLP 2026 implementation of Corpus2Skill: compile a document corpus into a navigable skill hierarchy that LLM agents explore at query time, with document lookup instead of a serving-time vector-search service.

Description provenance: Original source

Open canonical source

Measured: Oct 18, 2026 · 45 observations

Domains & Uses

Domains
Agents, skills and MCP
Uses
Agent skills, Retrieval-augmented generation (RAG)

Installation

git clone https://github.com/dukesun99/Corpus2Skill ~/.claude/skills/Corpus2Skill

Check at the source

Compatibility and prerequisites must be checked in the official documentation.

https://github.com/dukesun99/Corpus2Skill

Classifications

Nature
Tool

Measurement history

Comparable readings are grouped by signal and unit.

Stars

96stars · Oct 18, 2026

+13 (+15.7 %)since Sep 2, 2026

Sep 2, 2026 : 83 starsSep 4, 2026 : 84 starsSep 5, 2026 : 84 starsSep 6, 2026 : 85 starsSep 7, 2026 : 85 starsSep 8, 2026 : 85 starsSep 9, 2026 : 85 starsSep 10, 2026 : 85 starsSep 11, 2026 : 85 starsSep 12, 2026 : 87 starsSep 13, 2026 : 87 starsSep 14, 2026 : 87 starsSep 15, 2026 : 88 starsSep 16, 2026 : 88 starsSep 17, 2026 : 89 starsSep 18, 2026 : 89 starsSep 19, 2026 : 89 starsSep 20, 2026 : 89 starsSep 22, 2026 : 90 starsSep 23, 2026 : 90 starsSep 24, 2026 : 90 starsSep 25, 2026 : 91 starsSep 26, 2026 : 91 starsSep 27, 2026 : 92 starsSep 28, 2026 : 92 starsSep 29, 2026 : 92 starsSep 30, 2026 : 92 starsOct 1, 2026 : 92 starsOct 2, 2026 : 92 starsOct 3, 2026 : 92 starsOct 4, 2026 : 93 starsOct 5, 2026 : 93 starsOct 6, 2026 : 93 starsOct 7, 2026 : 93 starsOct 8, 2026 : 93 starsOct 9, 2026 : 93 starsOct 10, 2026 : 93 starsOct 11, 2026 : 94 starsOct 12, 2026 : 94 starsOct 13, 2026 : 94 starsOct 14, 2026 : 94 starsOct 15, 2026 : 94 starsOct 16, 2026 : 94 starsOct 17, 2026 : 95 starsOct 18, 2026 : 96 stars
Exact values · 45 readings
MeasuredstarsChange
96+1
95+1
940
940
940
940
940
94+1
930
930
930
930
930
930
93+1
920
920
920
920
920
920
92+1
910
91+1
900
900
90+1
890
890
890
89+1
880
88+1
870
870
87+2
850
850
850
850
850
85+1
840
84+1
83·