npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@losi-ai/core

v0.1.0

Published

Core context engine for Losi Context Bank — LLM adapter interface, context snapshots, memory and governance config. Portable context for any LLM.

Readme

@losi-ai/core

The context engine at the heart of Losi Context Bank. Portable context for any LLM — switch models freely, your context stays.

npm install @losi-ai/core

What's in here

  • LLMAdapter — the interface every model adapter implements.
  • LosiContext — holds memory + governance config and the active adapter. switchAdapter() swaps the model while keeping the context snapshot.
  • MemoryConfig / ContextSnapshot — declare what context to include; the snapshot is what gets injected and persisted.
  • GovernanceEngine / GovernanceConfig — in-memory spend caps, rate limits, blocked actions, data scopes, kill switch, audit log.
  • ContextBinding — the interface @losi-ai/nexus and @losi-ai/spaces implement.
  • Storage — serializeSnapshot / deserializeSnapshot and a pluggable SnapshotStore (MemorySnapshotStore included) for persisting context across sessions.
  • Resilience — withRetry (exponential backoff + jitter) and FallbackAdapter (try a chain of providers, fail over transparently).
  • Knowledge graph — KnowledgeGraph of nodes + typed edges with optional bi-temporal validity (addNode/addEdge/neighbors/renderGraph), plus temporal reasoning (queryValidRelations/invalidateEdge/supersede). Carried inside ContextSnapshot and rendered into the prompt.
  • Provenance & auditing — annotateProvenance (why context was included) and diffSnapshots (what changed between runs).

Example

import { LosiContext } from "@losi-ai/core";
import { OpenAIAdapter } from "@losi-ai/openai";

const ctx = new LosiContext({
  adapter: new OpenAIAdapter({ apiKey: process.env.OPENAI_API_KEY! }),
  memory: { crm: true, bookings: true },
  governance: { spendCapUsd: 5, rateLimitPerMinute: 60 },
});

const res = await ctx.complete("What's on my plate today?");
console.log(res.content);

Remember & recall

High-level memory verbs (familiar from Mem0/Cognee), backed by the same portable snapshot — so what you remember persists across sessions and model switches:

await ctx.remember("Acme Corp signed a 2-year contract in Q3.", { scope: "crm" });
await ctx.remember("The user prefers concise answers.");

const context = await ctx.recall("What do I know about Acme?");
// → rendered context (sections + relationships) ready to inject

remember() persists automatically when a SnapshotStore is configured; recall() refreshes from bindings and returns the injectable context.

Zero runtime dependencies. MIT licensed.