GitNova Daily discoveries — November 3, 2026 · 16:00 MSK

NVIDIA's Python framework that builds a semantic ontology graph over enterprise metadata, giving LLM agents trusted metric and entity definitions for reliable text-to-SQL.

Useful for

• Deploy the local stack via Docker Compose and connect a Databricks or Snowflake database for NL queries
• Generate an ontology from metadata and reconcile conflicting metric definitions across systems
• Serve context to agents via MCP, CLI or API with queries validated against certified definitions

Before using

• License: Apache-2.0
• Installation and requirements have not been verified; review the README before running

Explore project

Complete logs of every experiment from the blog post 'LLMs can understand and write neural network weights', including prompts, model replies, sandboxed code runs, evaluations and the networks read or written.

Useful for

• Reproduce an experiment from the post using the recorded prompts and run.json settings
• Inspect the model's reasoning summaries and sandboxed code output for a specific turn
• Reuse the evaluation scripts and answer keys to score weight-space tasks

Before using

• No license is listed in the GitHub data; check the usage terms
• Installation and requirements have not been verified; review the README before running

Explore project

An open-source toolkit for Spec-Driven Development that turns specifications into executable artifacts and guides the process from requirements to implementation with any AI coding agent.

Useful for

• Bootstrap a Spec-Driven Development workflow in a new project via specify init and /speckit-* commands
• Assess a raw idea through the intake→research→define→shape→decide flow with the assess extension
• Run a scoped assess→fix→test bug-fixing cycle using the bug extension

Before using

• License: MIT
• Installation and requirements have not been verified; review the README before running

Explore project