Outreach Deliverability Mcp

io.github.closermethodPublished by GitHub account closermethodv1.0.2Updated Oct 24, 2026

Channel-safety for cold outreach: volume limits, SPF/DKIM/DMARC, spam-trigger auditing.

VerifiedSTDIODesktop onlyCommunication & CollaborationBusiness & Commerce

Overview

AI-generated overview

A local MCP server that gives an assistant cold-outreach channel-safety guidance: volume limits, sender setup, spam-trigger audits and campaign diagnostics.

What you can do
Retrieve per-channel rules for email, LinkedIn, Instagram DM and X DM: safe daily volumes, account setup, warmup, kill signals and legal basics. Get a ten-step cold-email infrastructure checklist covering domains, SPF, DKIM, DMARC, mailboxes, warmup, list verification and monitoring. Audit message copy for spam triggers with a 0-100 risk score. Interpret open, reply, bounce and acceptance benchmarks against observed values. Diagnose which layer is failing (infrastructure, list or message) and the fix order. Return the whole library as one payload for full agent context.
When to use it
Use it when planning or reviewing a cold-outreach campaign and you want the assistant to check sending volumes, sender authentication setup and copy before sending. It is advisory only and does not send anything.
Requirements
Runs as a local process over stdio, installed from the npm package outreach-deliverability-mcp, so Node.js is needed. No accounts, API keys, environment variables or headers are declared. Desktop only.
Before you install
It returns rules, checklists and audits rather than sending messages, so it does not itself contact recipients. The README states that platform limits and filtering behavior change frequently and that its values are conservative practitioner ranges, not platform guarantees, and that legal notes are orientation, not legal advice. Treat its outputs as guidance to verify, not as authoritative policy.

Installation

In SourceWeft

  1. Open Outreach Deliverability Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Outreach Deliverability MCP

The channel-safety layer for cold outreach. Volume discipline, sender setup, spam auditing — so good messages actually arrive.

Disclaimer. Platform limits and filtering behavior change frequently; values are conservative practitioner ranges as of last verification, not platform guarantees. Legal notes (CAN-SPAM / GDPR / PECR / CASL) are orientation, not legal advice.

Why This Exists

Most outbound fails before the copy is ever read. Burned domains, throttled LinkedIn accounts, and spam-foldered sends kill campaigns that any message audit would have passed. AI agents make this worse: they can generate 500 perfect emails an hour, and sending them is exactly how you lose the domain.

This MCP is the layer that keeps the channel alive. It does NOT send anything — it returns rules, checklists, and audits that an agent consults BEFORE and DURING a campaign.

The core diagnostic it encodes: deliverability problems masquerade as copy problems. A 1% reply rate usually isn't a bad hook — it's a foldered send. Fix order: infrastructure → list → message.

6 Tools

ToolWhat it returns
get_channel_rulesSafe daily volumes, account setup, warmup protocol, kill signals, personalization floor, and legal basics per channel: email, LinkedIn, Instagram DM, X DM
get_sender_setup_checklistThe 10-step cold-email infrastructure checklist: lookalike domains, SPF, DKIM, DMARC, mailboxes, warmup, list verification, tracking, compliance, monitoring
audit_spam_triggersCopy audit: 30+ flagged phrases, all-caps, exclamations, link count, attachments, money symbols, fake Re:/Fwd: — with a 0-100 risk score
interpret_benchmarkHealthy/warning ranges for open rate, reply rate, bounce rate, LinkedIn acceptance, DM reply — with an optional direct read of your observed value
diagnose_campaignGiven your observed metrics, names the failure layer (infrastructure vs list vs message) and the fix order
get_full_packThe complete library as one payload for full agent context

Sample Use

typescript
// Campaign feels dead. Copy problem or channel problem?mcp.call("diagnose_campaign", {  channel: "email", open_rate: 31, reply_rate: 0.8, bounce_rate: 4.2});// Returns: failure layer = LIST. Bounce 4.2% > 3% is actively damaging the// domain; open 31% says sends are being foldered. Fix infrastructure and// list BEFORE touching the copy.
mcp.call("audit_spam_triggers", { text: "URGENT: Don't miss this LIMITED TIME offer!!! Click here..." });// Returns: HIGH RISK, phrase + structural hits itemized.

Works With

Pair with outbound-engine-mcp: that one makes the message worth sending; this one makes sure it lands.

Built By

Elisabeth Hitz — 10+ years of B2B enterprise sales experience across ad-tech, SaaS, media, and global hiring. Now building MCP servers for the AI agent ecosystem.

License: MIT

Source: README.md at commit 58c6214

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.2LatestSep 16, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.