Vantage

Created By
Cavendo-AI3 months ago
Vantage is a personal signal store for AI-native work. A place to capture and query the things you notice across the web. A post on X. A thread on LinkedIn. An insight from an article. A competitor move. Instead of losing it across tabs, screenshots, and notes, you save it -- and use it later.
Overview

Vantage

Track what you're seeing. Use it later.

Vantage is a personal signal store for AI-native work — a place to capture and query the things you notice across the web.

A post on X. A thread on LinkedIn. An insight from an article. A competitor move. Instead of losing it across tabs, screenshots, and notes, you save it — and use it later.


One memory layer across all your AI tools

Your AI tools don't share memory.

  • Claude doesn't know what you saw on X
  • ChatGPT doesn't know what you read yesterday
  • Your code tools don't know your research

Vantage fixes that.

Save signals once, and access them anywhere you have an MCP-compatible client — across Claude, Cursor, and other supported tools.


How it works

  1. Connect Vantage to your AI (via MCP)
  2. Save signals directly from your workflow
  3. Ask questions on top of what you've saved

Example:

"Save this as a signal"

"What have I saved about pricing?"

"Show me signals from Lemkin"


What Vantage is (and isn't)

  • Not a scraper — you choose what matters
  • Not a feed — no noise or algorithmic clutter
  • Not a knowledge base — no system to maintain

Vantage is a simple way to keep track of what you're seeing — so your AI can actually use it later.


Hosted + open source


Quick Start

Option 1: Use the hosted version (fastest)

Sign up at vantage.cavendo.ai. Connect to Claude or another supported client. Start capturing.

Takes about 2 minutes. No install required.

Option 2: Run locally

npm install
npm run dev

# If 3020 is already in use, choose another port
PORT=YOUR_PORT npm run dev

# Generate your first API key
curl -X POST http://localhost:3020/api/auth/keys \
  -H 'Content-Type: application/json' \
  -d '{"name": "my-key"}'

The first API key can be created from localhost without auth. If you're bootstrapping a fresh remote deployment, set VANTAGE_BOOTSTRAP_TOKEN on the server and send the same value in the X-Vantage-Bootstrap-Token header on that first request. The API key is returned as data.key in the JSON response.

Capture your first signal:

curl -X POST http://localhost:3020/api/signals \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer vtg_YOUR_KEY_HERE' \
  -d '{
    "signalType": "post",
    "platform": "x",
    "content": "Every SaaS company will need an agent layer within 18 months.",
    "topics": ["AI Agents", "SaaS"],
    "importance": "high"
  }'

MCP Server

Vantage includes an MCP server with 13 tools for use with Claude, Claude Code, Cursor, or any MCP-compatible client.

If you want to use the bundled stdio MCP server directly, install its dependencies too:

cd mcp-server
npm install
cd ..

Claude Code

claude mcp add vantage --transport http "http://localhost:3020/mcp" --header "Authorization: Bearer vtg_YOUR_KEY"

For manual testing, note that /mcp is a streamable HTTP endpoint. MCP clients handle the response format automatically, but ad hoc curl calls need to be prepared for stream-oriented responses rather than a simple JSON-only request/response flow.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "vantage": {
      "command": "node",
      "args": ["/path/to/vantage/mcp-server/src/index.js"],
      "env": {
        "VANTAGE_API_URL": "http://localhost:3020",
        "VANTAGE_API_KEY": "vtg_YOUR_KEY_HERE"
      }
    }
  }
}

MCP Tools

ToolWhat it does
capture_signalSave a post, article, quote, or discussion. Auto-creates sources and topics.
search_signalsFull-text search with filters — platform, topic, date range, importance.
list_recent_signalsLatest signals feed.
get_signalFull detail on a specific signal.
manage_sourceCreate or update a person/account you're tracking.
list_sourcesSee everyone you're tracking with signal counts.
manage_topicCreate or update a topic/theme.
list_topicsAll topics with signal counts.
manage_collectionGroup signals into named collections.
set_business_contextUpload strategy docs for comparison.
analyze_signalsAssemble signals + context for analysis.
get_dashboardSummary — counts, top topics, top sources, highlights.
save_analysisPersist an analysis for later reference.

Integrations

  • OpenClaw community skill: vantage-shared-intel — capture signals, search prior intel, and generate shared digests against the same Vantage dataset from OpenClaw.

REST API

Most API requests require Authorization: Bearer vtg_YOUR_KEY.

Exceptions:

  • GET /health
  • POST /api/auth/keys for first-key bootstrap from localhost, or from a remote client that supplies a valid X-Vantage-Bootstrap-Token
MethodEndpointDescription
POST/api/signalsCreate a signal
GET/api/signalsList signals with filters
GET/api/signals/search?q=queryFull-text search
GET/api/signals/feedChronological feed
GET/api/signals/:idSignal detail
POST/api/sourcesCreate a source
GET/api/sourcesList sources
POST/api/topicsCreate a topic
GET/api/topicsList topics
POST/api/collectionsCreate a collection
POST/api/contextsUpload business context
GET/api/dashboard/summaryDashboard summary
POST/api/auth/keysGenerate API key

Tech Stack

ComponentTechnology
RuntimeNode.js 20+ / Express / ESM
DatabaseSQLite via better-sqlite3
ValidationZod
AuthAPI key (vtg_ prefix), SHA-256 hashed
MCP@modelcontextprotocol/sdk (stdio + HTTP)

Environment Variables

VariableDefaultDescription
PORT3020Server port
NODE_ENVdevelopmentEnvironment
DATABASE_PATH./data/vantage.dbSQLite file path
CORS_ORIGIN*CORS allowed origin
VANTAGE_BOOTSTRAP_TOKENunsetAllows first API key creation from a non-localhost client when sent as X-Vantage-Bootstrap-Token

License

MIT — use it however you want.


Built by Jonathan Arehart at Cavendo AI.

Server Config

{
  "mcpServers": {
    "vantage": {
      "command": "node",
      "args": [
        "/path/to/vantage/mcp-server/src/index.js"
      ],
      "env": {
        "VANTAGE_API_URL": "http://localhost:3020",
        "VANTAGE_API_KEY": "vtg_YOUR_KEY_HERE"
      }
    }
  }
}
Project Info
Created At
3 months ago
Updated At
2 months ago
Author Name
Cavendo-AI
Star
-
Language
-
License
-
Category

Recommend Servers

View All
Docwand

14 hours ago
//beforeyouship — LLM Cost Modeling From Your Editor
@Indiegoing

Query realistic LLM cost models without leaving your editor. beforeyouship models the **true monthly cost** of an LLM app architecture — retries, prompt caching, batch discounts, infra overhead, and 3×/10× growth — across GPT-5.x, Claude, Gemini, DeepSeek, and more. Not a token calculator: a planning tool for the design phase, before you commit to a stack. **No API key needed to try it** — demo mode covers the six free-tier models. A Pro key from [beforeyouship.dev](https://beforeyouship.dev) unlocks the full 18-model catalog. ## What you can ask - "How much will a RAG chatbot cost at 10,000 requests/day?" - "Compare Claude Haiku vs Gemini Flash pricing for my workload" - "What's the cheapest model for a multi-step agent at scale?" - "Show me current per-token prices for Anthropic models" ## Tools ### `estimate_cost` Full cost model for an architecture at a given usage level. Returns Naive / Realistic / Worst Case monthly cost per model, 3×/10× growth scenarios, and an opinionated recommendation with reasoning. ### `get_model_prices` Current per-1M-token pricing — input, output, cached input, batch — with context windows and staleness metadata. ### `list_archetypes` Seven preset architecture patterns (simple chatbot, chatbot with history, RAG pipeline, multi-model router, coding assistant, document processor, multi-step agent) used as starting points for estimates. ## Setup **Claude Code:** ​```bash claude mcp add --transport http beforeyouship https://beforeyouship.dev/api/mcp ​``` **Cursor / other clients** — add a remote server: ​```json { "mcpServers": { "beforeyouship": { "type": "streamable-http", "url": "https://beforeyouship.dev/api/mcp" } } } ​``` Add an `Authorization: Bearer bys_...` header with a Pro key for the full catalog. ## Try it > Estimate the monthly cost of a RAG pipeline at 10,000 requests/day

14 hours ago