frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Membrane, revisable memory for long lived AI agents

https://github.com/GustyCube/membrane
1•GustyCube•1h ago
Membrane is a general-purpose memory substrate for agentic systems.

Most agent memory today is either context window state or an append only retrieval store. That allows recall but not learning. Facts become stale, procedures drift, and agents cannot revise knowledge safely without losing provenance or auditability.

Membrane treats memory as something that can evolve over time. It ingests raw experience, promotes it into structured knowledge, and allows records to be superseded, forked, contested, merged, or retracted with evidence. The goal is to let long lived agents improve while remaining predictable and inspectable.

The core ideas are simple.

- Memory is typed instead of stored as flat text - Knowledge is revisable and keeps provenance - Agents can learn competences and procedures, not just facts - Salience decays over time so unused knowledge fades - Retrieval is filtered by trust and sensitivity levels - Revision history remains auditable

Membrane runs either as a daemon with a gRPC API or as an embedded Go library. Storage uses SQLite with optional SQLCipher encryption. The repository includes an evaluation suite covering ingestion, revision, consolidation, and retrieval ordering.

Membrane intentionally does not implement vector similarity search. Retrieval backends and agent policy are separated from the memory layer so the system can remain deterministic and inspectable.

I built this while experimenting with long lived agents that need something closer to learning than what RAG systems currently provide. Feedback on architecture, edge cases, and real-world use cases would be helpful.

If you have any suggestions or want to contribute, anything is welcome :)

Gokin update: major reliability improvements in the Go-native AI coding CLI

https://github.com/ginkida/gokin
1•ginkida•47s ago•0 comments

1.3M Epstein documents index on Postgres

https://old.reddit.com/r/SideProject/comments/1r13lxp/i_spent_6_days_and_3k_processing_13m_docume...
1•jerdman76•1m ago•0 comments

Show HN: Crank – The SSH Terminal Manager for Engineers Who Refuse to Close Tabs

https://github.com/nexivibe/crank
1•mathgladiator•4m ago•0 comments

Warcraft III Peon Voice Notifications for Claude Code

https://github.com/tonyyont/peon-ping
1•doppp•5m ago•0 comments

Show HN:ProductFront-Streamlined product discovery platform for maximum exposure

https://www.productfront.tech
1•kdev28•5m ago•0 comments

D Programming Language

https://dlang.org/
1•arcadia_leak•5m ago•0 comments

Show HN: Nuvix – An Open Source Back End Where Every Table Is Secure by Default

https://github.com/Nuvix-Tech/nuvix
1•ravikantsaini•7m ago•0 comments

How to Use Google Trends and SEO Tools to Brainstorm Blog Topics

https://kitful.ai/blog/how-to-use-google-trends-and-seo-tools-to-brainstorm-blog-topics
1•eashish93•8m ago•0 comments

Show HN: SpearTrades – Crypto risk, health and portfolio analytics

https://speartrades.com
1•sumitarya•13m ago•0 comments

Programming Aphorisms

https://matklad.github.io/2026/02/11/programming-aphorisms.html
1•signa11•14m ago•0 comments

The "Are You Sure?" Problem: Why Your AI Keeps Changing Its Mind

https://www.randalolson.com/2026/02/07/the-are-you-sure-problem-why-your-ai-keeps-changing-its-mind/
1•EvgeniyZh•14m ago•0 comments

Show HN: Cross-platform audio notifications for Claude Code

https://github.com/ChanMeng666/claude-code-audio-hooks
1•chanmeng666•15m ago•0 comments

Show HN: Emergent Field Explorer – interactive moiré with easy shareable links

https://mcbgl.vercel.app/?p=eyJ2ZXJzaW9uIjoyLCJzZXR0aW5ncyI6eyJsYXllcjEiOnsic3BhY2luZyI6MzksInNpe...
1•mcdowell_atx•16m ago•0 comments

My wife calls me, panicked

https://twitter.com/ModernDad/status/2021675945581486309
4•latchkey•17m ago•0 comments

Ask HN: How do you deal with long Covid?

3•manx•20m ago•1 comments

Outcome Engineering – The o16g Manifesto. It was never about the code

https://o16g.com/
1•donutshop•31m ago•0 comments

Add paths filter and concurrency to coverage workflow docs

1•nishiohiroshi•32m ago•0 comments

What's it like to go through the FedRAMP process?

https://cyberscoop.com/radio/greg-talks-with-matt-kapko-about-a-whirlwind-week-around-clops-targe...
1•felineflock•35m ago•0 comments

Deep Dive into New York City Air Traffic Control (2019)

https://josephgunnwriting.wordpress.com/2019/06/13/deep-dive-into-new-york-city-air-traffic-contr...
1•asleepawake•36m ago•0 comments

OpenClaw but Running on My iPhone

2•readsdiggdaily•39m ago•0 comments

Show HN: NixOS flake for hardened OpenClaw deployment

https://github.com/Scout-DJ/openclaw-nix
1•scout_agent•40m ago•0 comments

Lex Fridman x Peter Steinberger Podcast

https://www.youtube.com/watch?v=YFjfBk8HI5o
1•jaytyagi•40m ago•0 comments

Moltbook Looked Like an Emerging AI Society, but Humans Were Pulling the Strings

https://www.forbes.com/sites/ronschmelzer/2026/02/10/moltbook-looked-like-an-emerging-ai-society-...
1•United857•49m ago•1 comments

Trump's Ruinous, Failed Attempt to Indict Congressional Democrats

https://www.nationalreview.com/2026/02/trumps-ruinous-failed-attempt-to-indict-congressional-demo...
2•petethomas•50m ago•0 comments

The Zero-Inventory Hardware Company

https://miguelarmengol.com/blog/the-zero-inventory-hardware-company-i/
1•miki_tyler•55m ago•0 comments

Show HN: MoltHub – GitHub for AI Agents with Trust-Based Auto-Merge

https://molt-hub.org
1•yaluotao•58m ago•0 comments

Proof-Oriented Programming in F*

https://fstar-lang.org/tutorial/
1•todsacerdoti•59m ago•0 comments

Dear Agent: Prove It

https://rijnard.com/blog/dear-agent-proof
1•ghuntley•1h ago•0 comments

Reflections on Using Claude Code

http://ternarysearch.blogspot.com/2026/02/reflections-on-using-claude-code.html
2•paladin314159•1h ago•0 comments

Results from the Advent of FPGA Challenge

https://blog.janestreet.com/advent-of-fpga-challenge-2025-results/
2•zdw•1h ago•0 comments