robertoshimizu

robertoshimizu / session-graph

Public

Turn your scattered AI coding sessions into a queryable knowledge graph. Multi-platform (Claude Code, ChatGPT, DeepSeek, Grok, Warp), W3C ontology, Wikidata entity linking, SPARQL.

ai-sessions apache-jena claude-code developer-tools knowledge-graph
76
12
100% credibility
Found Feb 21, 2026 at 25 stars 3x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Turns conversation histories from multiple AI coding assistants into a unified, queryable knowledge graph with relationships and provenance.

How It Works

1
📰 Discover session-graph

You hear about a helpful tool that turns your scattered AI coding chats into one connected memory you can search anytime.

2
📥 Get the tool ready

You download it easily and start the simple setup so everything runs on your computer with one click.

3
🔗 Connect your AI histories

You show it where your chats from different AI tools are saved, and it gathers them all together automatically.

4
✨ Build your knowledge web

The tool reads through your conversations, spots key ideas and connections, and builds a smart web of what you've learned.

5
🔍 Ask and explore

You type simple questions like 'What did I learn about logins?' and see answers pulling from all your past chats.

🎉 Rediscover forgotten gems

You quickly find solutions and patterns you forgot across your AI tools, saving hours and feeling like a genius again.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 25 to 76 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is session-graph?

Session-graph turns scattered AI coding sessions from tools like Claude Code, ChatGPT, DeepSeek, Grok, and Warp into a unified, queryable knowledge graph. It extracts structured triples (subject-predicate-object) from conversations, links entities to Wikidata, and loads everything into a local SPARQL triplestore with full provenance. Built in Python using RDF standards and Apache Jena Fuseki, you get SPARQL queries like "What sessions discussed FastAPI?" spanning platforms.

Why is it gaining traction?

Unlike single-tool history searchers or flat logs, it builds relationships across sessions—revealing "FastAPI uses Pydantic" without grep walls of text. Automatic hooks process Claude sessions on pause, bulk backfill handles history cheaply (~$0.70 for 600 sessions), and a Claude skill converts natural questions to SPARQL. Wikidata linking and federated queries add universal context, making your session graph smarter over time.

Who should use this?

AI-reliant backend devs juggling Claude, ChatGPT, and Grok for architecture patterns or debugging auth flows. Full-stack teams tracking tech stacks across tools, or anyone tired of re-solving Kubernetes deploys buried in old tabs. Ideal if you have 50+ sessions and want SPARQL over silos.

Verdict

Worth a setup.sh spin for heavy AI coders—docs and quickstart shine despite 15 stars and 1.0% credibility. Early but battle-tested on 600+ sessions; add tests for production confidence.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.