synthos-akne-bridge

Diagnoses and repairs bidirectional connections between Synthos papers and the AKNE knowledge graph.

6|1|Updated May 11, 2026
One-click install
npx skills add https://github.com/yakeworld/Synthos --skill synthos-akne-bridge-yakeworld
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: synthos-akne-bridge
Source: https://github.com/yakeworld/Synthos/tree/main/skills/private/extended/research-tools/synthos-akne-bridge
Command: npx skills add https://github.com/yakeworld/Synthos --skill synthos-akne-bridge-yakeworld

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires networkx, jieba, sentence-transformers, and includes references (resource) components.

What problem does it solve? It fixes broken or missing links between the Synthos paper pipeline and the AKNE knowledge graph, eliminating orphaned papers and skills so knowledge flows bidirectionally across the graph. ## Core Features & Use Cases - Bridge Diagnostics: Runs akne-query.sh bridge to audit connection status, orphan counts, and bidirectional path availability before any repair. - Graph Repair Operations: Connects orphaned papers and skills to domain categories and Wiki concepts, creates reverse edges (source_category, category_paper, concept_paper), and normalizes paper directories into 01-manuscript/06-references/07-quality structure. - Content Injection & Vectorization: Injects content_summary, content_hash, and heading metadata into source nodes and backfills vectors.db so semantic search covers all 1145 source files. - Use Case: After adding new Synthos papers, run the bridge audit to find orphaned nodes, map them to AKNE categories by domain keywords, create reverse edges, and verify orphans=0. ## Quick Start Run the AKNE bridge diagnostic on my knowledge graph and repair any orphaned Synthos papers or skills until the audit reports zero orphans.

Frequently Asked Questions about synthos-akne-bridge

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

FAQPage Schema
How do I fix orphaned papers in a knowledge graph?

Run the bridge audit first to diagnose connection status, then manually map orphaned papers to domain categories using keyword matching and add paper_source_domain edges. Verify by re-running the audit until orphans equal zero.

How do I create bidirectional edges in a NetworkX knowledge graph?

Explicitly create reverse edge types such as source_category, category_paper, and concept_paper alongside the forward edges. Note that source_category_membership runs category-to-source, so the reverse source-to-category edge must be added separately.

Why does find_related fail with too many values to unpack?

The find_related method returns 3-tuples of neighbor, relation chain, and metadata, not 4 values. Unpacking with four variables raises ValueError; use three-variable unpacking or check the API documentation.

Can I import the knowledge graph library inside a venv without sentence-transformers?

No. The venv Python 3.11 environment lacks sentence-transformers, so importing the query engine there raises ModuleNotFoundError. Route all graph queries through the terminal using system Python 3.12 instead.

Why do multi-word English queries return no results in graph search?

The bag-of-words graph search cannot match multi-word English queries. Use the full query mode with jieba tokenization and multi-token substring matching, or fall back to vector search against the embeddings database.