JSPM

@glgio/know-me

0.1.1
  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 23
  • Score
    100M100P100Q54770F
  • License MIT

AI package for developer portfolios — chat widget + MCP server, queryable profile from a single know-me.yaml.

Package Exports

  • @glgio/know-me/core
  • @glgio/know-me/mcp
  • @glgio/know-me/widget

Readme

know-me

AI package for developer portfolios. Drop a know-me.yaml next to your site and you get two things:

  • Chat widget — a React component that lets visitors talk to an AI that knows your background.
  • MCP server — a Model Context Protocol endpoint so AI agents (Claude, Cursor, etc.) can query your profile programmatically.

One npm package, one source of truth (know-me.yaml), edge-deployable.

Install

npm install @glgio/know-me

You'll also need an Anthropic API key for the chat. The MCP endpoint works without one.

1. Write your know-me.yaml

Drop it at the root of your site project. Minimum:

identity:
  name: Your Name
  title: What you do

Full schema in examples/know-me.yaml. Sections:

  • identity (required: name)
  • summary
  • experience[] — company, role, start, end?, stack?, highlights?
  • skills — primary[], secondary[], learning[], notes?
  • projects[]
  • education[]
  • links[]
  • availability — status: open | selective | closed
  • contact

2. Add the chat endpoint

Edge function (Vercel example, src/pages/api/chat.ts):

import { parseProfile, buildContext } from "@glgio/know-me/core";
import yamlSource from "../../../know-me.yaml?raw";
import Anthropic from "@anthropic-ai/sdk";

export const config = { runtime: "edge" };

const profile = parseProfile(yamlSource);
const system = buildContext(profile);
const client = new Anthropic({ apiKey: process.env.ANTHROPIC_API_KEY! });

export default async function handler(req: Request) {
  const { messages } = await req.json();
  const stream = await client.messages.stream({
    model: "claude-haiku-4-5-20251001",
    max_tokens: 1024,
    system,
    messages,
  });

  const encoder = new TextEncoder();
  const body = new ReadableStream({
    async start(controller) {
      for await (const event of stream) {
        if (
          event.type === "content_block_delta" &&
          event.delta.type === "text_delta"
        ) {
          controller.enqueue(encoder.encode(event.delta.text));
        }
      }
      controller.close();
    },
  });
  return new Response(body, { headers: { "Content-Type": "text/plain" } });
}

3. Drop in the widget

import { ChatWidget } from "@glgio/know-me/widget";

export default function Page() {
  return (
    <ChatWidget
      title="Chat with my profile"
      suggestions={["What did you ship last?", "Are you open to roles?"]}
    />
  );
}

4. Add the MCP endpoint

// src/pages/api/mcp.ts
import { handleMcpRequest } from "@glgio/know-me/mcp";
import { parseProfile } from "@glgio/know-me/core";
import yamlSource from "../../../know-me.yaml?raw";

export const config = { runtime: "edge" };
const profile = parseProfile(yamlSource);

export default function handler(req: Request) {
  return handleMcpRequest(req, { profile });
}

Tools exposed:

  • get_profile — full structured profile
  • get_skills — skills grouped
  • match_role — heuristic match against a job description

API reference

@glgio/know-me/core

parseProfile(source: string): KnowMeProfile
buildContext(profile: KnowMeProfile, opts?: { persona?: string }): string

@glgio/know-me/widget

<ChatWidget
  endpoint="/api/chat"
  title="Chat with my profile"
  greeting="Ask me anything…"
  suggestions={["…"]}
  mode="floating" // or "inline"
  theme={{ accent: "#2563eb" }}
/>

@glgio/know-me/mcp

createKnowMeServer({ profile }): { dispatch(req): Promise<JsonRpcResponse | null> }
handleMcpRequest(req: Request, { profile }): Promise<Response>

Requirements

  • Node 18+
  • A host that runs edge functions (Vercel, Netlify, Cloudflare). GitHub Pages alone won't work.

License

MIT