Back to search
reachiz-mcp
SkillWorks with Claude CodeGitHub
Reachiz MCP server connector and agent skills: LinkedIn + email outreach, lead lists, enrichment and Autopilot plays.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/gsmmediaro--reachiz-mcp)HTML
<a href="/listings/gsmmediaro--reachiz-mcp"><img src="/api/listings/gsmmediaro--reachiz-mcp/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
plugin.json · Manifest · 14 lines
- {
- "$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
- "name": "reachiz-mcp",
- "version": "1.0.0",
- "description": "Reachiz MCP server connector and agent skills: LinkedIn + email outreach, lead lists, enrichment and Autopilot plays.",
- "author": {
- "name": "Reachiz",
- "url": "https://app.reachiz.com"
- },
- "homepage": "https://reachiz.com/mcp",
- "repository": "https://github.com/gsmmediaro/reachiz-mcp",
- "license": "MIT",
- "keywords": ["mcp", "outreach", "linkedin", "leads", "reachiz"]
- }
README.md · README · 90 lines
- # Reachiz MCP
- Reachiz is a LinkedIn + email outreach platform and a credit-paid people-data API (lead lists, finds, Autopilot, unibox). This repo holds the public manifests and agent skills for its hosted MCP server.
- **Endpoint:** `https://app.reachiz.com/mcp` — MCP Streamable HTTP, one POST per call (stateless).
- ## Connect
- ### OAuth
- The server is an OAuth 2.0 protected resource. Clients that support remote MCP OAuth discover everything they need:
- ```json
- {
- "mcpServers": {
- "reachiz": { "url": "https://app.reachiz.com/mcp" }
- }
- }
- ```
- An unauthenticated call returns `401` with `resource_metadata`, so the client follows the sign-in flow automatically.
- ### API key
- Alternatively, create a key in the Reachiz dashboard (Settings → API keys) and send it as a bearer token:
- ```json
- {
- "mcpServers": {
- "reachiz": {
- "url": "https://app.reachiz.com/mcp",
- "headers": { "Authorization": "Bearer <rz_live_...>" }
- }
- }
- }
- ```
- Key scopes mirror the REST API (`lists.read`, `leads.write`, `billing.read`, ...). `billing.*` needs a tenant-admin key.
- ### Per client
- - **Claude (claude.ai)** — Settings → Connectors → add `https://app.reachiz.com/mcp`, then complete the OAuth sign-in.
- - **ChatGPT** — Settings → Connectors → Advanced → Developer mode → create a connector with the URL above and OAuth.
- - **Cursor** — `.cursor/mcp.json` with the config block above, or open a `cursor://` install deeplink.
- - **Codex / raw config** — same JSON block in the client's MCP config.
- ## What it does
- `tools/list` returns a small default profile (22 tools): workspace/account reads, people and company search, list create/add/export, email/phone enrich and verify, credits packs and checkout, plus seven meta-tools: `get_started`, `search_tools`, `describe_tool`, `call_tool`, `post_quote`, `submit_feedback`, `report_missing_tool`.
- The full catalog (152 tools — every `/v1` route) is reachable through `call_tool`; find names with `search_tools` and schemas with `describe_tool`. Calling a non-default tool name directly returns "Invalid params" — wrap it in `call_tool`.
- ## Gate model
- Tools that send, spend credits, change access, or delete data refuse until you pass `confirm: true`, and their descriptions say so. Before spending, call `post_quote` with the tool name and arguments to get the max credit cost without running it.
- ## Examples
- **Find leads and enrich a list**
- ```
- post_searches { body: { title: "Head of Marketing", location: "Germany", maxResults: 25, listName: "DE marketing" }, confirm: true }
- post_lists_by_id_enrich { params: { id: 12 }, body: { what: "email" }, confirm: true }
- get_lists_by_id_export { params: { id: 12 } } → CSV text
- ```
- **Review an Autopilot play, then launch**
- ```
- get_autopilot → plays[]
- call_tool { name: "get_autopilot_plays_by_id", arguments: { params: { id: 1 } } }
- call_tool { name: "post_autopilot_plays_by_id_approve", arguments: { params: { id: 1 }, confirm: true } }
- call_tool { name: "post_autopilot_plays_by_id_launch", arguments: { params: { id: 1 }, body: { when: "now" }, confirm: true } }
- ```
- **Check credits before a spend**
- ```
- get_credits → { balance: 50, ... }
- post_quote { name: "post_lists_by_id_enrich", arguments: { params: { id: 3 }, body: { what: "email" } } }
- → { credits: 3, gate: { confirmRequired: true } }
- ```
- ## Errors
- Tool failures come back as `isError: true` with a text message ending in a stable code in parentheses: `invalid_request`, `not_found`, `forbidden`, `conflict`, `unavailable`, `rate_limited`. A gate refusal reads "Not run: this tool … Ask the user to approve it, then call again with confirm: true".
- ## Legal
- Endpoint: https://app.reachiz.com/mcp · Product: https://reachiz.com/mcp · Privacy: https://app.reachiz.com/privacy-policy · Terms: https://app.reachiz.com/terms-of-service · Contact: https://app.reachiz.com/contact
Version history
v1.0.011/22/2026
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.