AMP – open-source memory server for AI agents (MCP, SQLite, D3.js)
Hi HN, I’m Akshay. I built AMP because I was tired of my AI agents having "amnesia" the moment I closed the terminal. Like many of you, I use Claude/Cursor daily. RAG is great for searching documentation, but it’s terrible for continuity. It chunks text blindly, losing the narrative. When I asked my agent "Why did we decide to use FastAPI last week?", it would hallucinate or just give me generic pros/cons because the specific context of our decision was lost in a vector soup. So I decided to build a proper *Hippocampus* for my agents. *What is it?* AMP is a local-first memory…
What it does
In the maker’s words, at launch
Hi HN, I’m Akshay. I built AMP because I was tired of my AI agents having "amnesia" the moment I closed the terminal. Like many of you, I use Claude/Cursor daily. RAG is great for searching documentation, but it’s terrible for continuity. It chunks text blindly, losing the narrative. When I asked my agent "Why did we decide to use FastAPI last week?", it would hallucinate or just give me generic pros/cons because the specific context of our decision was lost in a vector soup. So I decided to build a proper *Hippocampus* for my agents. *What is it?* AMP is a local-first memory server that sits between your agent and its LLM. It implements a "3-Layer Brain": 1. *STM*: A scratchpad for what we are doing right now. 2. *LTM*: Consolidated facts and insights (promoted from STM). 3. *Galaxy Graph*: A force-directed knowledge graph (d3.js) that physically links related concepts. *The Tech Stack* * *Protocol*: Built natively for the *Model Context Protocol (MCP)*. If you use Claude Desktop or Cursor, it works out of the box. * *Backend*: Python, FastAPI, SQLite (no docker needed). * *Search*: Hybrid (Keyword + Vector) using FastEmbed. * *Viz*: A 60fps local dashboard to actually see your agent's brain growing. *Benchmarks* I didn't want just another vector wrapper, so I focused heavily on retention. I benchmarked it against Mem0 (a popular alternative) on the LoCoMo dataset. * *AMP*: 81.6% Recall (Context-First) * *Mem0*: 21.7% Recall (Extraction-First) It turns out that preserving the narrative (who/what/when) before summarizing is key to avoiding "I don't know" answers. I’d love to hear your thoughts on the architecture. Does your agent need a hippocampus? Repo: https://github.com/akshayaggarwal99/amp
Does the same job
all alternatives →
AMP by CanyonTechs AI26d ago · app.canyontechs.ai · ▲91AI agents that act. Automation that delivers.
CrewAI AMP - Agent Management PlatformOct 2025 · ▲29Orchestrate AI agents with trust, scale & simplicity
- BABuilt an Obsidian plugin that rephrases your writing without takin overJun 2026 · rephrasethis.co · ▲5
Writing is hard, and it's tempting to just let AI do the whole thing So I built an Obsidian plugin that keeps AI in its place Highlight a sentence, get some options, pick the one you like Sharpens your writing instead of automating it
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, December 2025
the whole month →- GP
Life & fun · Dec 2025 · dosaygo-studio.github.io
- JG
Hi everyone! My name's Luke and I made the original Jmail here alongside Riley Walz. We had a ton of friends collaborate on building out more of the app suite last night in lieue of DOJ's "Epstein files" release. Please AMA!
Life & fun · Dec 2025 · jmail.world
PlanEat AI▲733AI turns your health goals into a 7-day menu & grocery list
AI · Dec 2025 · planeatai.com
- 2G
Community, All the HN belong to you. This is an archive of hacker news that fits in your browser. When I made HN Made of Primes I realized I could probably do this offline sqlite/wasm thing with the whole GBs of archive. The whole dataset. So I tried it, and this is it. Have Hacker News on your device. Go to this repo (https://github.com/DOSAYGO-STUDIO/HackerBook): you can download it. Big Query -> ETL -> npx serve docs - that's it. 20 years of HN arguments and beauty, can be yours forever. So they'll never die. Ever. It's the unkillable static archive of HN and it's…
Dev tools · Dec 2025 · hackerbook.dosaygo.com
