Urgent.News

What's breaking now, across thousands of outlets.

AI

handoff: letting parallel Claude Code sessions know why each other changed things

Built solo for Graph Hacks (WeMakeDevs × FalkorDB). Code: https://github.com/adityailab/graph-agent Demo video: https://youtu.be/16bKhR6Q08c The problem I run Claude Code in several terminals at once: a frontend session, a backend session, a deploy session. They are good at their own work and blind to each other. The backend session can read the frontend's diff, but a diff says what changed,…

Graph Hacks and FalkorDB collaborated on a project that lets multiple Claude Code sessions communicate about their work. Each session can read and write to a shared graph that records the intent behind each action. This includes the original request, the decision made, and any dependencies between roles. The graph is built using GraphRAG-SDK and contains entities like Pipeline, Job, and DatabaseTable.

Claude Code sessions can search the graph using meaning vectors and typo-tolerant full-text search, as well as insights from FalkorDB's algorithms like WCC, PageRank, and label propagation. A coordinator agent reminds roles about outstanding needs and sends digests. The system runs entirely on the user's Claude Code login, without an API key.

A demo showed three Claude Code sessions working together to build a small shop application, with the coordinator reminding the backend to write missing tests after 95 minutes of inactivity. The knowledge graph was live, with sessions interacting through messages, reads, and needs.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at dev.to →

More in AI

GraphGuard: Building a Graph-Powered AI Agent for Fraud Investigation with FalkorDB

Track: Track 1 — Agents That Act on Connected Data Event: Graph Hacks 2026 by WeMakeDevs and FalkorDB Introduction AI agents are becoming increasingly capable of answering questions, analyzing…

  • GraphGuard is an AI prototype for fraud investigation
  • Utilizes FalkorDB graph database for data modeling
  • Highlights include account analysis, transaction tracing

Free AI softphone that records, transcribes and prompts

Most of my working day happens on calls: customer calls through the office PBX, and meetings in Zoom, Teams and Meet. The tools for those two worlds never talk to each other. Softphones just dial.

  • AI-Softphone records, transcribes, and prompts SIP calls locally
  • Pre-installed assistants handle subtitle generation, translation, objection handling
  • App supports local models via Ollama and cloud services with API keys

More from Sunday 18 October →