TVB Wiki – Connectome‑Based Whole‑Brain Modeling

A growing knowledge base for computational neuroscience, focusing on connectome‑based whole‑brain modeling, neural mass models, neuroimaging modalities, and related software.

Structure

• raw/papers/ – arXiv papers as markdown (metadata + abstract)
• raw/articles/ – other sources (bioRxiv, medRxiv, GitHub READMEs)
• entities/ – pages for software, authors, labs, datasets
• concepts/ – pages for methods, modalities, topics
• comparisons/ – comparative analyses
• queries/ – saved queries and search patterns
• meta/ – internal state (entity counts, timestamps, lint logs)
• scripts/ – Python scripts for automated updates and linting

Automated Updates

The wiki is updated automatically via two cron jobs:

1. Hourly research (scripts/hourly_update.py) – fetches new papers from arXiv, updates entity/concept pages, rebuilds the index.
2. Daily lint (scripts/daily_lint.py) – checks for orphan pages and broken internal links.

Cron logs are saved to ~/.hermes/cron/output/. The wiki's own log is at log.md.

Usage

Open this directory as an Obsidian vault (set OBSIDIAN_VAULT_PATH to this path).

Browse index.md for a categorized list of all pages.

Schema

The SCHEMA.md file defines the tag taxonomy used for categorizing pages.

Adding Content Manually

1. Drop a new paper in raw/papers/ as markdown.
2. Run the ingest script to update entity/concept pages.
3. Or just edit pages directly.

AI-Powered Content Enrichment (Feynman)

This wiki integrates Feynman research skills for automated content generation:

• /lit - Literature reviews for concept pages
• /deepresearch - Deep investigations for complex topics
• alpha - AlphaXiv paper search and Q&A
• /audit - Paper-code consistency checks

Quick Enrichment

# Analyze wiki for placeholder content
python3 scripts/enrich_wiki.py

# Run full enrichment pipeline
python3 scripts/feynman_wiki_pipeline.py --full

# Then execute generated Feynman commands
/lit "NEST simulator features architecture"

See docs/FEYNMAN_INTEGRATION.md for detailed workflows.

Development

To extend the automation:

• Edit scripts/hourly_update.py to add new sources (bioRxiv, medRxiv, GitHub).
• Adjust entity extraction logic in the same script.
• Modify SCHEMA.md to refine the tag taxonomy.
• Use Feynman skills (/lit, /deepresearch) to fill placeholder content.

Credits

Built with Hermes Agent's llm‑wiki and obsidian skills.

Static Site

This wiki is also a static site built with MkDocs + Material theme. The site/ directory contains the generated HTML. A GitHub Actions workflow automatically deploys the site to GitHub Pages on every push to main.

Last seeded: 2026‑04‑20 with 122 arXiv papers.