# ekas-io/open-sales-stack [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/ekas-io/open-sales-stack  
**GitHub Stars:** 2  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/ekas-io-open-sales-stack

## Description
Collection of B2B sales intelligence MCP servers. Includes website analysis, tech stack detection, hiring signals, review aggregation, ad tracking, social profiles, financial reporting and more for AI-powered prospecting by Ekas

## Claude Desktop Quick Installation
Heuristic fallback — verify the package name and runner against the repository README before running it. Uses `npx` (confidence: low):

```json
"mcpServers": {
  "open-sales-stack": {
    "command": "npx",
    "args": ["-y","ekas-io-open-sales-stack"],
    "env": {
      "LLM_PROVIDER": ""
    }
  }
}
```

**Requires environment variables:** `LLM_PROVIDER` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What ekas-io/open-sales-stack MCP server does

The ekas-io/open-sales-stack MCP server collection provides separate local MCP packages for B2B sales research. Its ready packages cover website intelligence, technology detection, social profiles, hiring activity, and advertising. Website research extracts product, pricing, team, and company details from a site. Technology research identifies tools such as CRM, marketing automation, analytics, chat, and support software from page source. Other ready packages inspect LinkedIn profiles and posts, job listings, and active campaigns in LinkedIn and Meta ad libraries.

Additional packages are listed but marked as in progress. These include review, funding, news, public-company financial reporting, firmographic, and GitHub intelligence. The repository also includes skills for workflows such as lead qualification, prospect research, LinkedIn reconnaissance, and cold-email personalization.

## How it works

Each MCP operates independently, so you can add only the research packages relevant to your workflow or install the full collection. Claude reads the available tool descriptions and selects calls based on the request. The packages can also be chained: a company-research prompt might combine website findings with detected technologies, hiring signals, social information, and advertising activity.

The ekas-io/open-sales-stack MCP server collection runs on the user's machine and works with public web data. LLM-based extraction requires an API key from OpenAI, Anthropic, or Google Gemini. The README states that no additional API keys are needed beyond the LLM provider key, although social-intel offers optional LinkedIn authentication for some workflows.

## Setup and configuration

The repository requires Python 3.10 or newer and an LLM API key. Setup is performed by cloning the repository, running `bash scripts/setup.sh`, and completing the provider and key prompts. `bash scripts/verify.sh` checks the installation. The setup process writes configuration to `.env`; the default model can later be changed through the `LLM_PROVIDER` value, with supported formatting documented in `.env.example`.

Use `bash scripts/add-to-claude.sh --all` to add the MCPs. The script targets Claude Code when its CLI is available and otherwise targets Claude Desktop. Use `--desktop` or `--code` to select the destination explicitly, or pass individual package flags such as `--website-intel`, `--social-intel`, and `--hiring-intel`.

LinkedIn authentication for social-intel can be skipped initially, completed through a browser login, or configured with credentials for headless login. Company scraping works without a LinkedIn login according to the setup notes.

## Tools and capabilities

The ekas-io/open-sales-stack MCP server collection includes these ready packages:

- `website-intel` for structured information from company websites
- `techstack-intel` for technologies detected in page source
- `social-intel` for LinkedIn company pages, people profiles, and posts
- `hiring-intel` for roles across listed job sites and direct career pages
- `ad-intel` for campaigns, creatives, and targeting signals from LinkedIn and Meta ad libraries

The included high-inbound-volume qualification skill uses five research signals and can save results to Apollo. Skills provide workflow instructions; the MCPs provide the research data.

## Limitations and notes

Not every listed package is ready. Review, funding, news, financial-reporting, firmographic, and GitHub intelligence are marked as in progress in the repository table, so their availability and behavior should be verified before relying on them. The collection is designed around public web research and does not describe private-data access. A supported LLM API key is required for extraction, and LinkedIn-specific authentication may be needed for some social research.

_Full upstream README: https://allmcps.com/mcp/ekas-io-open-sales-stack/readme_