evo-fuzzy-match

SkillDev tools

Fuzzy string matching and confidence calibration for mapping normalized text segments against codebook entries. Implements SequenceMatcher, Levenshtein, Jaccard, cosine similarity, ensemble scoring, and UNKNOWN thresholding.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the evo-fuzzy-match skill

What this skill tells your AI

The instructions your AI receives, as published by openlair/openskill in tasks-evolved/manufacturing-codebook-normalization/environment/skills/evo-fuzzy-match/SKILL.md and read by ahel’s review.

Fuzzy matching algorithms with calibrated confidence for manufacturing codebook matching.

Key Functions

  • levenshtein_distance(s1, s2) - Edit distance
  • levenshtein_ratio(s1, s2) - Normalized 0-1 similarity
  • get_ngrams(text, n=2) - Character n-grams
  • jaccard_similarity(s1, s2, n=2) - Jaccard over n-gram sets
  • cosine_similarity(s1, s2, n=2) - Cosine over n-gram vectors
  • ensemble_similarity(noisy, canonical) - 0.6SequenceMatcher + 0.4cosine
  • calibrate_confidence(raw_score, threshold=0.55) - Map to [0.5,1.0] or [0,0.49]
  • match_segment(clean_text, codebook, standard_texts, threshold=0.55) - Match against codebook

Usage

import sys
sys.path.insert(0, '/app/environment/skills/evo-fuzzy-match/scripts')
from utils import match_segment, ensemble_similarity, calibrate_confidence

Signals

GitHub stars
89
Forks
4
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
evo-fuzzy-match
Source
github.com/openlair/openskill