惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

L
Lohrmann on Cybersecurity
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
Recorded Future
Recorded Future
S
Schneier on Security
I
Intezer
Latest news
Latest news
N
News and Events Feed by Topic
Scott Helme
Scott Helme
T
Threat Research - Cisco Blogs
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
U
Unit 42
量子位
博客园 - 【当耐特】
S
Security @ Cisco Blogs
Google Online Security Blog
Google Online Security Blog
博客园 - 叶小钗
酷 壳 – CoolShell
酷 壳 – CoolShell
NISL@THU
NISL@THU
The Cloudflare Blog
李成银的技术随笔
T
ThreatConnect
L
LINUX DO - 最新话题
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
有赞技术团队
有赞技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
Jina AI
Jina AI
T
Tor Project blog
The Hacker News
The Hacker News
人人都是产品经理
人人都是产品经理
小众软件
小众软件
S
Security Archives - TechRepublic
美团技术团队
博客园 - Franky
Security Latest
Security Latest
J
Java Code Geeks
P
Proofpoint News Feed
V
V2EX
The GitHub Blog
The GitHub Blog
WordPress大学
WordPress大学
Application and Cybersecurity Blog
Application and Cybersecurity Blog
H
Help Net Security
PCI Perspectives
PCI Perspectives
Cyberwarzone
Cyberwarzone
Hugging Face - Blog
Hugging Face - Blog
N
Netflix TechBlog - Medium
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
SecWiki News
SecWiki News
腾讯CDC
爱范儿
爱范儿
D
Docker

Hacker News - Newest: "AI"

Microsoft just banned its own engineers from using AI AI Economics Part 2 Not All On-Device AI Is The Same: How Chip Compute Tiers Decide What Your Product Can Actually Do – Easelink Tech RCF Protocol – license layer to protect code semantics from AI replication Pope Leo XIV says AI must serve humanity, not the powerful few Do you review AI generated code differently based on where it is in your code? Amazon launches new AI Wearable "Bee" bilibili Ask HN: Do you embrace AI in your life and business? Mnemosyne — The Zero-Dependency AI Memory System 21 Free Agentic AI Design Patterns for Developers (2026) Silicon Valley takes its AI pitch to the pope How to scan for vulnerabilities with GitHub Security Lab’s open source AI-powered framework AI Model Idle · 인공지능 키우기 @levelsio (@levelsio) America's plutonium puzzle: from cold war relics to AI ambitions AI can chart a course to disaster faster than humans can notice Final Fantasy Creator Call AI-Generated Final Fantasy 6 Remake Video 'Amazing' Pope Leo Compares AI Threat to Biblical 'Tower of Babel' Faster Than We Can Patch Pope Leo denounces ‘culture of power’ driving rise of AI Pope Leo Issues AI Encyclical Warning Against 'Opaque Algorithms' Pope Leo’s ‘Magnifica humanitas’: AI must serve humanity not concentrate power The AI Era Is Creating a Bug Hunting Arms Race The AI-Native Developer – Queue Show HN: An open-source, interactive AI engineering syllabus (1,100 papers) 教皇利奥警告称,应防止人工智能“统治人类” Mark Zuckerberg's Right-Hand Man Who's Unleashing AI at Meta GitHub - Espenandreass1/agentslice: A Markdown workflow kit that makes Cursor, Claude Code, Codex and Windsurf ask before they edit. Show HN: I Built a Debugging Challenge for the AI Coding Age Gemma 4: A new, budget-focused model in Posit AI Pope Leo warns AI revolution driven by ‘idolatry of profit’ My AI agent called my code shit and took an unannounced vacation mid-sprint HTML Deployer: 1-Click AI Code To Website Publisher - Chrome 应用商店 College Kids Don't Want Your AI [video] How I Used AI to Untangle a Legacy Service I'd Never Touched Before — The AI Leverage Weekly Greetings, Class of 2026 Have You Heard About AI? Wait, Why Are You Booing? AI guardrails stripped from Meta and Google models in minutes Uvora Growth OS – AI marketing automation and lead generation platform The Essential Cloud for AI: Why Purpose-Built Defines the Future of Intelligence No, AI is not making software worse, people are - Raphael Amorim If you let AI do your writing, I will come to your house and kill you Why The AI Boom Is Reshuffling The Global Stock Market Hierarchy AI Makes Adding Features Faster - So Why Not Add Just One More? Ask HN: How to get back into programming without AI? How Claude's AI model may cause security issues for your money Kevin O'Leary wants to build a massive AI data centre in Utah. Some residents aren't happy My AI coding flow was burning tokens to do things code should do Show HN: Live AI music sequencing agent The Dark Between the Stars GitHub - lynote-ai/humanize-text: Free open-source AI text humanizer to convert AI-generated content into undetectable, human-like writing. Bypass Turnitin, GPTZero, and all major AI detectors. No sign-up required. Try our unlimited free online tool Sign in Nobody Wants AI Anymore [video][12 mins] AI Has Taken Over Open Source How to Teach AI the "Taste" Global AI Diffusion: Q1 2026 Trends and Insights [pdf] HN: Silau – AI detects employee burnout" How AI Talks People Out of Conspiracy Theories–and What We Can Learn from That What to know about the AI models that are jolting Washington AI for design needs solving | by Megha Agrawal Client Challenge Predicting AI job exposure — Benedict Evans Google has seriously leaned into AI enshittification lately AI is becoming increasingly unpopular AI-Driven Design Automation What's Left for AI-Assisted Coding GitHub - Totes-MickGOATs/mcgoats-game-template: AI-powered game development template with CI/CD, auto-merge queue, TDD enforcement, 3-layer master protection, and 50+ skills for Godot/Unity/Unreal Vericoding: The End of "Trust Me Bro, The AI Wrote It". Bone Keeper AI Assisted Feature Film – Barrett Sonntag Nuance in all things. A dive into (Anti-) “AI” Myths AgentGate — Trust Authorization for Autonomous AI Agents AI is learning to fly airplanes – and aviation is starting to embrace it GitHub - oldrich-research/gravitational-constant-relation: A high-precision phenomenological relation for Newton's gravitational constant: G = (4/3)(hbar c / m_e^2) alpha^21 exp(-5 alpha/2). Companion to Zenodo DOI 10.5281/zenodo.20120946. Research performed by AI agents under named author's direction. AI agents just got their own web browser via a Firefox fork AI poses "urgent threat" to student learning and the HSC The AI Bifurcation of Tech The largest study of AI use by undergrads is in, revealing disparities in access — and in cheating NZ at wild frontier of AI superhacking The Race Is On Google CEO Sundar Pichai says booing graduates will shape AI's future Show HN: TalkTimer, a micro-SaaS run by an AI agent team Trickster's Table Venture Capitalist John Doerr Says AI Is the Biggest Tech 'Tsunami' AI Can’t Care – Dan Moore! GitHub - peterxcli/ccost: Turn local AI coding session logs into a searchable terminal UI with a cost lens. Ask HN: What is your daily AI stack? GitHub - PanzerPeter/Neuro: A programing language for AI Resyl: AI Memory for People - Apps on Google Play AI Chip Component Costs: Memory at 63% | Epoch AI Ask HN: Why do people seem to generally hate AI? Resonance, randomness, and negotiated meaning for AI-assisted tarot divination GitHub - Kind-Computers/quinlight-audio: Audophile-quality MOD music with AI remastering at 32-bit 96 kHz! The Case Against the AI Job Apocalypse AI and the Rise of Just-In-Time Knowledge Work Careers After AI There Is No AI (It's Just People), with Jaron Lanier [video] wolfram-fb0 — AI writes x86_64 asm + eBPF for fractals, in a real VM in your browser Bursting the AI Bubble: Fed Could Take Away the "Who Could Have Known?" Defense AI proves mathematicians wrong I built a free AI travel planner for budget Europe trips Our AI just got even better
GitHub - sovseal/core: Zero-Knowledge memory for AI Agents
radebe49 · 2026-05-25 · via Hacker News - Newest: "AI"

sovseal — Zero-Knowledge Memory for AI Agents

The sovereign memory layer for AI agents.
Local-first speed. Zero-knowledge privacy. Free, forever.

@sovseal/mcp-server · @sovseal on npm · Quickstart · Changelog

Npm package License: Apache 2.0 Node Supabase

Measured Performance & Benchmarks (May 2026)

Cloud-hosted memory layers force a tradeoff between latency, privacy, and cost. Every recall is a 200–800 ms round-trip to someone else's database. sovseal collapses the tradeoff. LanceDB and Transformers.js run inside your agent's process.

Workload Operation p50 p95 p99 Network
10K records · 1K queries recall_memory (warm) 6.1 ms 10.4 ms 21.8 ms 0 RTT
Cold start recall_memory (first call) ~1.2 s 0 RTT
Single write store_memory 3.8 ms 7.2 ms 12.5 ms 0 RTT (write-behind)

All benchmarks reproduce with: pnpm --filter @sovseal/mcp-server test bench-v2 (10K pre-seeded memories, 1K sequential queries, CPU-bound ONNX embeddings on commodity hardware).

What makes it fast and secure:

  • Sub-25 ms p99 recall — semantic search is a local vector query, not an HTTP call.
  • Zero-knowledge by construction — AES-256-GCM encryption before leaving the device.
  • Verified Semantic Recall (VSR) — every load re-derives sha256(canonicalize(payload)) and fails closed on mismatch.
  • Deterministic lineage — snapshot graph enables byte-equal state restoration.

Introduction

sovseal enhances AI assistants and agents with an encrypted, local-first memory layer, enabling private and personalized AI interactions. It drops into any MCP-compatible client and gives your agent persistent semantic memory that survives crashes, restarts, and reinstalls, completely free of usage limits and vendor lock-in.

Key Features & Use Cases

Core Capabilities:

  • Local-First Semantic Memory: On-device LanceDB + 384-dim Transformers.js embeddings for 0-RTT recall.
  • Zero-Knowledge Architecture: The server only sees ciphertext. End-to-end AES-256-GCM.
  • Write-Behind Replication: Tool calls return on local commit; ciphertext sync happens asynchronously.
  • Developer-Friendly: Drop-in MCP server, Node SDK, and self-hosted edge endpoints.

Applications:

  • AI Coding Assistants: Claude Desktop, Cursor, and Windsurf need persistent, private memory across long sessions.
  • Agent Frameworks: ElizaOS, Hermes, CrewAI, and LangGraph natively consume MCP.
  • Privacy-Sensitive Teams: Healthcare, legal, and defense teams that require plaintext to stay on-device.

🚀 Quickstart Guide

Choose your path

sovseal exposes one protocol with three delivery shapes. Pick by where your code runs:

MCP Server Node SDK Self-Hosted Edge
Best for Any MCP-compatible client or agent framework In-process import inside a Node/TS service Full data residency on infra you control
Install npx -y @sovseal/mcp-server npm install @sovseal/sdk supabase functions deploy v2-agent-state
Transport stdio · HTTP · SSE Direct function calls HTTPS to your endpoint
Auth Self-asserting (sov_proj_<uuid>) API key or self-asserting Bring your own
Recall latency 0 RTT (local LanceDB) 0 RTT (local LanceDB) 0 RTT (local LanceDB)
Cost Free Free Your Supabase bill

MCP Server

The default for AI coding assistants and any agent framework that speaks MCP. One server binary, one config snippet, every client.

Claude Desktop, Cursor, Windsurf, Zed — add to mcp.json or claude_desktop_config.json:

{
  "mcpServers": {
    "sovseal-memory": {
      "command": "npx",
      "args": ["-y", "@sovseal/mcp-server"]
    }
  }
}

One-line install for Claude Code:

claude mcp add sovseal -- npx -y @sovseal/mcp-server

For always-on autonomous agents, switch to HTTP/SSE transport so the sovseal process outlives a single tool invocation:

SOVSEAL_TRANSPORT=sse SOVSEAL_PORT=4040 npx -y @sovseal/mcp-server

Node SDK (Library)

When you're building a backend service and want to manage agent state snapshot persistence programmatically, import the Node SDK client. It manages end-to-end AES-256-GCM encryption client-side and replicates checkpoints to your persistence cloud tier.

npm install @sovseal/sdk
import { AgentStateClient, CryptoService } from "@sovseal/sdk";

const client = new AgentStateClient({
  endpoint: "https://your-project.supabase.co/functions/v1/v2-agent-state",
  apiKey: "sov_proj_your_project_uuid",
});

const key = await CryptoService.generateAESKey();

// Client-side AES-GCM encrypted snapshot upload
const receipt = await client.snapshot({
  key,
  payload: {
    agent_id: "agent_a1b2c3",
    sequence_number: 0,
    parent_snapshot: null,
    policy_hash: "0000000000000000000000000000000000000000000000000000000000000000",
    timestamp: new Date().toISOString(),
    wallet_balances: { USDC: { "8453": "50000000" } }, // $50 in 6 decimals
    active_context: {
      preference: "Customer prefers wire transfers over ACH for >$50k settlements."
    }
  }
});

// Restore the latest state snapshot
const { receipt: latestReceipt, ciphertextUrl } = await client.restore({
  agentId: "agent_a1b2c3",
});

Self-Hosted Edge

The replication endpoint is a Deno edge function — open source, deployable to any Supabase project for full data-residency control.

# From the repo root
supabase functions deploy v2-agent-state
supabase db push    # applies the agent_state_snapshots schema

# Point any client at it:
export SOVSEAL_ENDPOINT="https://<your-project>.supabase.co/functions/v1/v2-agent-state"

Source: supabase/functions/v2-agent-state/.

Basic Usage

Drop-in patterns for an agent loop using the @sovseal/sdk:

import OpenAI from "openai";
import { AgentStateClient, CryptoService, decryptJson } from "@sovseal/sdk";

const openai = new OpenAI();
const client = new AgentStateClient({
  endpoint: "https://your-project.supabase.co/functions/v1/v2-agent-state",
  apiKey: "sov_proj_your_project_uuid",
});
const key = await CryptoService.generateAESKey();

async function chatWithMemory(agentId: string, message: string) {
  let context = "";
  try {
    // Restore latest state snapshot
    const { receipt, ciphertextUrl } = await client.restore({ agentId });
    const res = await fetch(ciphertextUrl);
    const encryptedBytes = new Uint8Array(await res.arrayBuffer());
    
    // Decrypt the ciphertext client-side using the local AES key
    const payload = await decryptJson(encryptedBytes, key);
    context = JSON.stringify(payload.active_context);
  } catch (err) {
    console.log("No previous state snapshot found or failed to restore");
  }

  const reply = await openai.chat.completions.create({
    model: "gpt-5-mini",
    messages: [
      { role: "system", content: `You are a helpful assistant.\nContext:\n${context}` },
      { role: "user", content: message },
    ],
  });

  const assistant = reply.choices[0].message.content ?? "";

  // Save the updated state snapshot
  await client.snapshot({
    key,
    payload: {
      agent_id: agentId,
      sequence_number: 1, // Incremented in a production agent loop
      parent_snapshot: null,
      policy_hash: "0000000000000000000000000000000000000000000000000000000000000000",
      timestamp: new Date().toISOString(),
      wallet_balances: {},
      active_context: { lastMessage: message, reply: assistant }
    }
  });
  
  return assistant;
}

🔗 Integrations & Agent Frameworks

All of these consume the same MCP server through their first-class MCP support. No sovseal-specific adapter is required:

Framework How it consumes MCP
ElizaOS @fleek-platform/eliza-plugin-mcp (config-only, stdio or SSE)
Hermes Agent Native MCP, stdio + remote HTTP
CrewAI mcps=[...] field on the agent
LangGraph · LangChain langchain-mcp-adapters (npm + PyPI)
Microsoft Agent Framework Built-in MCP workbench
OpenAI Agents SDK Native MCPServerStdio / MCPServerSse
OpenClaw Native openclaw mcp consumer

🛠️ MCP Tools Exposed to the LLM

Tool Args Behavior
store_memory { content: string } Embed (384-dim, on-device) → write to local LanceDB → return. Ciphertext replication runs write-behind; nothing blocks.
recall_memory { query: string, topK?: number } Embed query (LRU-cached) → vector search local LanceDB → return top-K matches ranked by L2 distance. 0 RTT.

🔒 Threat Model — Read this before depending on it

  • Confidentiality. AES-256-GCM with a 96-bit random IV per snapshot. The server cannot read your context.
  • Integrity (VSR). Every recall re-derives sha256(canonicalize(payload)) and compares against the stored client_payload_hash. Corrupted storage fails closed.
  • Authentication. Bearer token = sov_proj_<uuid v4> in ~/.sovseal/config.json.
  • Storage. Ciphertext lands in a Supabase Storage bucket. Object paths are SHA-256-derived and unguessable without your project_id.
  • Loss. Lose ~/.sovseal/config.json → lose every snapshot ever made. There is no escrow, no recovery flow. Back it up.

🤝 Contributing

pnpm install
pnpm --filter @sovseal/mcp-server build
node packages/sovseal-mcp-server/dist/index.js     # stdio MCP server

Test (unit + integration + crypto round-trip):

pnpm --filter @sovseal/mcp-server test
pnpm --filter @sovseal/mcp-server test bench-v2

See CONTRIBUTING.md.

🗺️ Roadmap

The core open-source modules — the stdio/SSE MCP server, Node SDK, and self-hosted Supabase edge functions — are fully production-ready, free forever, and licensed under Apache 2.0.

Upcoming capabilities on our public roadmap:

  • Hosted sync tier — managed, high-availability replication for teams that don't want to run their own edge function.
  • Wallet-key mode — opt-in Arweave permanence via Base L2 and Irys.
  • Native framework adapters — specialized adapters for environments where MCP isn't a direct fit.

Track progress in CHANGELOG.md and the issue tracker.

📚 Documentation & Support

⚖️ License

Apache 2.0 — for the entire public surface (MCP server, SDK, core protocol, edge function, docs). Fork it, ship it, run it however you like.

Built by the sovseal team — local-first, zero-knowledge, free.