match-kg-microbe

Align CHEBI-annotated ingredient sets to KG-Microbe medium nodes and populate kg_microbe_match.

10|3|Updated Jan 28, 2026
One-click install
npx skills add https://github.com/CultureBotAI/CultureMech --skill match-kg-microbe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: match-kg-microbe
Source: https://github.com/CultureBotAI/CultureMech/tree/main/.claude/skills/match-kg-microbe
Command: npx skills add https://github.com/CultureBotAI/CultureMech --skill match-kg-microbe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

CultureMech media recipes are hard to cross-reference with the KG-Microbe knowledge graph unless you can reliably assign the correct kg_microbe_match node ID for each recipe.

Core Features & Use Cases

  • Exact KG node identification: Matches each CultureMech record to the appropriate mediadive.medium:XXX KG-Microbe node using exact CHEBI ingredient set equality (concentrations ignored).
  • Traceability via kg_microbe_match: Populates the kg_microbe_match field with the matched node ID to enable downstream KG queries and validation against authoritative DSMZ-derived formulations.
  • Workflow flexibility: Supports full-repo batch matching, per-category enrichment, and single-recipe inspection for targeted debugging.

Quick Start

Run the KG matcher in dry-run mode across the repository to preview which media would be matched, then re-run the same command without the dry-run flag to write kg_microbe_match values.

Frequently Asked Questions about match-kg-microbe

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

FAQPage Schema
How do I match CultureMech media recipes to KG-Microbe node IDs?

Match CultureMech media recipes to KG-Microbe node IDs by aligning CHEBI-annotated ingredient sets to mediadive medium nodes using exact set equality. This process populates the kg_microbe_match field with the correct mediadive.medium:XXX identifier for downstream knowledge graph queries.

Can I run KG-Microbe matching on a single recipe instead of the whole repository?

Single-recipe inspection is supported alongside full-repo batch matching and per-category enrichment. This allows targeted debugging of individual mediadive medium records to verify CHEBI ingredient set equality before committing to a repository-wide run.

Do I need CHEBI term IDs before matching media to the KG-Microbe knowledge graph?

CHEBI term IDs must be added to ingredient sets before running the KG-Microbe matcher. The matching mechanism relies on exact CHEBI ingredient set equality, ignoring concentrations, to align CultureMech records with the appropriate mediadive medium nodes.

Why are my media records not matching to KG-Microbe nodes?

Matching fails when CHEBI-annotated ingredient sets do not achieve exact equality with mediadive medium solution sets. Concentrations are ignored during comparison, but the CHEBI term IDs themselves must perfectly align with the nodes defined in the KG-Microbe edges.tsv file.