# Saber

The Saber MCP server has tools available for creating company and contact signals, retrieving signals, managing lists and managing Saber settings.

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add saber/remote-mcp

# Browse available tools
npx -y @smithery/cli@latest tool list saber/remote-mcp

# Get full schema for a tool
npx -y @smithery/cli@latest tool get saber/remote-mcp company_signals.create_and_wait

# Call a tool
npx -y @smithery/cli@latest tool call saber/remote-mcp company_signals.create_and_wait '{}'
```

## Direct MCP Connection

Endpoint: `https://remote-mcp--saber.run.tools`

## Tools (22)

- `company_signals.create_and_wait` — Create a company signal and block until the result is ready — returns the answer in the same response (recommended over…
- `company_signals.list` — List company signals with optional filters for domain, company ID, date range, and status
- `company_signals.create` — Create a company signal asynchronously — returns immediately with a pending status; poll the returned ID or receive the…
- `company_signals.create_batch` — Create multiple company signals in batch — combines domains and questions using a Cartesian product; use templates for …
- `company_signals.get` — Get a company signal by ID — returns current status and AI-generated answer if completed
- `contacts.get_research_by_external_id` — Get contact research using an external system's ID (e.g. a HubSpot contact ID) and source name
- `contacts.create_research` — Start a contact research job — AI gathers insights from LinkedIn and other sources asynchronously; use contacts.get_res…
- `contacts.get_research` — Get contact research by ID — returns status and AI-generated insights if completed
- `contacts.search` — Search for contacts at a company via LinkedIn Sales Navigator — requires a LinkedIn Sales Navigator connection on the A…
- `contacts.create_signal_and_wait` — Create a contact signal and block until the result is ready — returns the answer in the same response (recommended over…
- `contacts.list_signals` — List contact signals with optional filters for LinkedIn profile URL and pagination
- `contacts.create_signal` — Create a contact signal asynchronously — returns immediately with a pending status; poll the returned ID or receive the…
- `contacts.get_signal` — Get a contact signal by ID — returns current status and AI-generated answer if completed
- `signal_templates.list` — List all signal templates for your organization
- `signal_templates.create` — Create a reusable signal template — templates define standard research questions that can be applied to many companies …
- `signal_templates.get` — Get a signal template by ID
- `signal_templates.update` — Update a signal template — creates a new version while preserving the template ID; omitted fields retain their previous…
- `signal_templates.delete` — Soft-delete a signal template — marks it as deleted but preserves it for historical tracking
- `signal_summaries.list` — List all AI-generated signal summaries for a domain, ordered by creation date (latest first)
- `signal_summaries.generate` — Generate an AI summary consolidating insights from all completed company signals for a domain
- `connectors.list` — List connectors and their connection status
- `credits.get_balance` — Get remaining credits balance — returns how many API credits your organization has left in the current billing period

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get saber/remote-mcp <tool-name>
```

## Resources

- `saber://capabilities` — A guide to all Saber MCP tools — what they do, when to use each one, and tips for best results.

## Prompts (3)

- `research-account` (domain, focus) — Research a company before a sales call — runs key signal questions and produces a structured sales brief
- `qualify-account` (domain, criteria) — Qualify a company against your ICP criteria and return a scored verdict with evidence
- `research-contact` (linkedin_url, company_domain) — Research a contact before a meeting — gathers LinkedIn insights and surfaces conversation openers