Glia is a 100% offline, open-source memory bridge. A Chrome extension auto-saves your web-based Claude/ChatGPT chats, while a native MCP server lets Cursor/Claude Code query those decisions locally from your shared SQLite database.
Hey Product Hunt! 👋
I'm Eshaan, the creator of Glia.
Like most developers working with AI, my daily workflow became deeply fragmented. I’d solve a complex architectural challenge or debug a tricky error using Claude.ai, ChatGPT, or Gemini in my browser. But the second I switched back to my local editor (like Cursor, Windsurf, or Claude Code), my IDE agent had absolutely no idea that conversation had happened.
I got tired of constantly copy-pasting code blocks, raw logs, and decision summaries back and forth. Existing memory solutions are almost all cloud-based SaaS, which felt like a massive privacy compromise for client codebases.
So I built Glia—a 100% local-first, zero-Docker memory bridge that connects your web chats directly to your local IDE.
The Architecture:
The Shared Database: Everything is stored offline on your local disk in a single SQLite database (~/.glia/graph.db). No telemetry, no cloud tracking, and your data never leaves your device.
The Web Sync: A lightweight Chrome extension securely intercepts prompts on 7 AI platforms (Claude, ChatGPT, Gemini, DeepSeek, Grok, Copilot, and Mistral) to save and index your technical decisions automatically.
The IDE integration: A native Model Context Protocol (MCP) server hooks into your IDE (Cursor, VS Code, Windsurf, Claude Code) so your coding agents can query or save memory.
Data Portability: Easily download any project session as a clean JSON file to sync manually to another machine or share with teammates offline.
Preemptive FAQs / Objection Handling:
Why not just use native codebase indexing in Cursor? Cursor indexes your files, and Claude Projects stores context for individual web sessions. But they are completely siloed. Claude Projects doesn't know what you coded in Cursor, and Cursor has no idea what you just solved in Claude.ai. Glia acts as the missing bridge that syncs context between them.
Isn't a local LLM in the background a major resource drain? No heavy 8B models are running in the background. Glia uses a lightweight embedding model (nomic-embed-text ~270MB via Ollama) and sqlite-vec for local vector search. It's extremely light on CPU and RAM.
How do you handle database write locks during embeds? We configured SQLite with Write-Ahead Logging (PRAGMA journal_mode=WAL;) and decoupled the embedding generation using an async background job queue. This ensures browser writes never block IDE reads.
Quick Start:
Setup takes less than a minute. You can spin up the local worker and register it with your IDE configs using:
npx glia-ai-setup
It's completely open-source (MIT). I'd love to hear your feedback, bugs, or feature suggestions. If Glia makes your development life easier, dropping a star on GitHub would mean the world to us! ⭐
What IDE or web platform would you like to see supported next?
This solves a problem I’ve started noticing myself while switching between ChatGPT, Claude and coding tools. A lot of useful context gets lost across chats, tabs and IDEs, so having a local memory bridge feels genuinely useful instead of just “AI for AI”.
Also like that it’s offline/open-source. Feels much safer and more practical for developer workflows where context and previous decisions actually matter.
nice! local-first plus the browser-to-IDE bridge is the gap most ai workflows leak into. congrats on shipping.
when the browser chat and the actual codebase contradict (chat assumes library x, codebase migrated off it last week), which wins? file-system recency is the safe answer, but it throws away conversational nuance the chat captured. tie-breaker, or human prompted? best of luck with your launch!
This is awesome! I built something similar, but took the opposite direction — no local setup, pure clipboard injection within certain LLM's.
This looks like an absolute game-changer for developer workflows! Bridging the gap between browser brainstorming and the IDE using an MCP server is brilliant, and keeping it 100% offline with a local SQLite database is a huge win for privacy.
Great work! The 7-platform Chrome extension surface is probably the trickiest bet here? Claude, ChatGPT, Gemini all rework their DOM constantly and each one breaks differently.
Really interesting approach to local memory. The SQLite + MCP combo is clean. Curious how you handle context relevance, when Cursor queries past decisions, how does it decide which memories are actually useful vs noise from older conversations?
@Glia The Hybrid RAG setup caught my eye immediately.. fusing sentence vectors, chunk vectors, and FTS5 keyword search together feels way more solid than what most memory systems are doing right now. I’ve been working on a Corrective RAG setup myself, and one of the biggest headaches was retrieval completely falling apart once the query got rephrased or drifted semantically from the original context. HyDE honestly feels like a really smart workaround for that.. generating a hypothetical answer first and then searching using that embedding instead of the raw query makes a lot of sense in practice.
What I’m curious about though is whether the synthetic embedding step adds noticeable latency during recall. In my experience even tiny delays start compounding very quickly once everything is happening inside an agent loop, especially when multiple retrieval passes are involved.
The shared SQLite bridge between the browser extension and MCP server is also honestly a really elegant design choice.. one database, two interfaces, no extra sync layer headaches. But I’d genuinely love to know how you’re handling write concurrency there. SQLite’s single-writer lock can get annoying fast, and if Cursor plus the browser extension both try writing context at the same time, does GLIA queue the writes internally or can one request fail silently? Feels like the kind of issue that would be super subtle to debug once an agent session is already running and actively mutating state.
Congrats on the launch! This is something I do feel daily: solve something in the Claude web app, switch to Claude Code in terminal, lose all that context. The MCP server + local SQLite combo is a great architectural bet for this.
Quick q for you: most of what happens in a Claude/ChatGPT chat is exploration, dead ends, half-formed ideas. How does Glia decide what to index as a meaningful technical decision vs noise? Is it post-hoc LLM extraction at save time, user-marked, or scored on whether the result actually got applied in code?
ok this does tackle a huge discomfort of mine. Good job & congrats on your launch
About Glia on Product Hunt
“Local-first AI memory bridge between browser chats and IDEs”
Glia launched on Product Hunt on May 20th, 2026 and earned 102 upvotes and 27 comments, placing #12 on the daily leaderboard. Glia is a 100% offline, open-source memory bridge. A Chrome extension auto-saves your web-based Claude/ChatGPT chats, while a native MCP server lets Cursor/Claude Code query those decisions locally from your shared SQLite database.
Glia was featured in Productivity (652.1k followers), Developer Tools (512.8k followers), Artificial Intelligence (469.1k followers) and GitHub (41.2k followers) on Product Hunt. Together, these topics include over 318.8k products, making this a competitive space to launch in.
Who hunted Glia?
Glia was hunted by Eshaan. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Glia stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Eshaan