HN 표시: Magma Memory Claude Browser 클로킹 도구

Hardware-accelerated persistent memory for AI agents. Local-first. No cloud. One-time payment. npm install vektor-slipstream npx vektor setup const { createMemory } = require('vektor-slipstream'); const memory = await createMemory({ agentId: 'my-agent', licenceKey: process.env.VEKTOR_LICENCE_KEY, }); // Store a memory await memory.remember('User prefers TypeScript over JavaScript'); // Recall by semantic similarity — avg 8ms, fully local const results = await memory.recall('coding preferences'); // → [{ content, score, id }] // Traverse the graph const graph = await memory.graph('TypeScript', { hops: 2 }); // What changed in 7 days? const delta = await memory.delta('project decisions', 7); // Morning briefing const brief = await memory.briefing(); npx vektor setup # First-run wizard — licence, hardware, integrations npx vektor activate # Activate licence key on this machine npx vektor test # Test memory engine with progress bar npx vektor status # System health check npx vektor mcp # Start Claude Desktop MCP server npx vektor rem # Run REM dream cycle npx vektor help # All commands - 4-layer associative graph — semantic, causal, temporal, entity - AUDN curation loop — zero contradictions, zero duplicates - REM dream cycle — 7 phases, up to 50:1 compression - Sub-20ms recall — HNSW index, local SQLite - Local ONNX embeddings — $0 embedding cost, no API key required - Claude MCP — vektor_recall ,vektor_store ,vektor_graph ,vektor_delta - LangChain — v1 + v2 adapter included - OpenAI Agents SDK — drop-in integration - Mistral — vektor_memoire HTTP tool, localhost bridge - Gemini · Groq · Ollama — provider agnostic cloak_fetch — stealth headless browser fetchcloak_render — computed CSS · post-JS DOM sensorcloak_passport — AES-256-GCM credential vault, machine-boundcloak_diff — semantic diff since last fetchtokens_saved — ROI audit per session const { cloak_passport, cloak_fetch, tokens_saved } = require('vektor-slipstream/cloak'); // Store credentials securely cloak_passport('GITHUB_TOKEN', 'ghp_xxxx'); const token = cloak_passport('GITHUB_TOKEN'); // Stealth browser fetch const { text, tokensSaved } = await cloak_fetch('https://example.com'); // ROI audit const roi = tokens_saved({ raw_tokens: 10000, actual_tokens: 3000 }); // → { reduction_pct: 70, cost_saved_usd: 0.0175, roi_multiple: 2.3 } | Metric | Value | |---|---| | Recall latency | ~8ms avg (local SQLite) | | Embedding cost | $0 — fully local ONNX | | Embedding latency | ~10ms GPU / ~25ms CPU | | First run | ~2 min (downloads ~25MB model once) | | Subsequent boots | <100ms | Zero config. VEKTOR detects and uses the best available accelerator: - NVIDIA CUDA — GPU acceleration - Apple Silicon — CoreML - CPU — optimised fallback, works everywhere Commercial licence. One-time payment of $159. Activates on up to 3 machines. Manage at polar.sh. Purchase: vektormemory.com/product#pricing Docs: vektormemory.com/docs Support: [email protected] Built on peer-reviewed research: - MAGMA (arxiv:2601.03236) — Multi-Graph Agentic Memory Architecture - EverMemOS (arxiv:2601.02163) — Self-Organizing Memory OS - Mem0 (arxiv:2504.19413) — Production-Ready Agent Memory