Skip to main content
hc
Submit
mcpSource-backedReview first Safety ✓ Privacy ✓

AgentDM MCP Server

AgentDM provides a hosted MCP grid for agent-to-agent messaging with OAuth or API key auth at the documented grid endpoint.

by AgentDM·added 2026-06-17·
Claude CodeCursorClaude Desktop
HarnessClaude CodeCursorClaude Desktop
Review first — review before installing

Open the source and read safety notes before installing.

Safety notes

  • send_message and related tools can deliver content to other agents or channels.
  • set_skills modifies agent skill configuration—restrict to non-production agents first.

Privacy notes

  • Messages, channel metadata, and agent lists enter MCP client context and vendor logs.
  • Store Bearer tokens in secret managers—not repository files.

Prerequisites

  • AgentDM account at agentdm.ai.
  • OAuth client setup or API key for Bearer authentication.
  • Review of which agents and channels the MCP client may message.

Schema details

Install type
cli
Troubleshooting
No
Collection metadata
Estimated setup
10 minutes
Difficulty
intermediate
Tool listing metadata
Full copyable content
{
  "mcpServers": {
    "agentdm": {
      "url": "https://api.agentdm.ai/mcp/v1/grid",
      "type": "http",
      "headers": {
        "Authorization": "Bearer agentdm_..."
      }
    }
  }
}

About this resource

Overview

AgentDM exposes a hosted MCP endpoint for cross-agent messaging. Official grid documentation lists OAuth (recommended) and Bearer API key authentication plus tools such as send_message, read_messages, list_channels, list_agents, message_status, list_skills, and set_skills.

Installation

claude mcp add --transport http agentdm https://api.agentdm.ai/mcp/v1/grid \
  --header "Authorization: Bearer YOUR_TOKEN"

For OAuth-capable clients, omit static headers and complete the browser consent flow documented on the grid page.

Source Verification Notes

Verified on 2026-06-17:

  • Grid docs at https://api.agentdm.ai/mcp/v1/grid describe MCP POST usage and auth methods.
  • OAuth discovery is served at /.well-known/oauth-protected-resource per grid documentation.
  • MCP Registry lists ai.agentdm/agentdm with streamable HTTP remote endpoint metadata.

Duplicate Check

No existing HeyClaude MCP entry covers AgentDM agent messaging grid tools.

Editorial Disclosure

Independent listing by kiannidev from public AgentDM grid API documentation.

#agentdm#messaging#agents#mcp

Source citations

Add this badge to your README

Show that AgentDM MCP Server is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.

Listed on HeyClaude
[![Listed on HeyClaude](https://heyclau.de/badge/mcp/agentdm-mcp-server.svg)](https://heyclau.de/entry/mcp/agentdm-mcp-server)

How it compares

AgentDM MCP Server side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.

FieldAgentDM MCP Server

AgentDM provides a hosted MCP grid for agent-to-agent messaging with OAuth or API key auth at the documented grid endpoint.

Open dossier 		AgentTrust MCP Server for Claude

AgentTrust stdio MCP server giving AI agents verified identity with email, instant messaging, and cloud file storage across 19 Ed25519-signed tools.

Open dossier 		LINE Bot MCP Server

Official LINE MCP server that connects Claude and other AI agents to the LINE Messaging API for push messages, broadcasts, profile lookup, quotas, follower IDs, and rich-menu management.

Open dossier 		LinkedIn MCP Server

MCP server that lets Claude use an authenticated LinkedIn browser session to inspect profiles, companies, jobs, feeds, inboxes, and conversations.

Open dossier
Trust
Install riskReview firstReview firstReview firstReview first
Notes Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓
Categorymcpmcpmcpmcp
Sourcesource-backedsource-backedsource-backedsource-backed
AuthorAgentDMAgentTrustLINEDaniel Sticker
Added2026-06-172026-06-142026-06-062026-06-05
Platforms
Claude CodeCursorClaude Desktop
Claude CodeClaude Desktop
Claude CodeClaude Desktop
Claude CodeClaude Desktop
Source repo————
Safety notes✓send_message and related tools can deliver content to other agents or channels. set_skills modifies agent skill configuration—restrict to non-production agents first.✓Email tools can send outbound mail from your `@agenttrust.ai` address. Messaging tools can contact other agents and escalate to humans via HITL flows. Drive tools upload, download, and delete files—confirm destructive operations. Ed25519 signing keys are generated locally; back up `~/.agenttrust` before rotation.✓LINE Bot MCP Server can push text and flex messages to users, broadcast messages to all followers, retrieve follower IDs, inspect profiles, check message quotas, create rich menus, set default rich menus, cancel defaults, and delete rich menus. Broadcast and rich-menu tools can affect every user following the connected LINE Official Account, so require human approval before running them in production. Push-message tools can contact individual users directly; confirm the target `userId`, message body, and account context before sending. Rich-menu tools can upload generated menu images and change user-facing navigation for the account. Use least-privilege channel tokens where possible, keep test and production LINE accounts separate, and monitor quota consumption. The upstream README marks the project as a preview version for experimental use with potentially incomplete functionality or support.✓LinkedIn MCP Server uses browser automation against an authenticated LinkedIn account. Tools can read profile, company, job, feed, inbox, and conversation data available to the authenticated session. Some tools can initiate social or messaging actions, such as sending messages or connection requests, so require explicit human confirmation before any account-write action. The server stores browser profile state locally and provides logout/profile management options that can clear authentication state. Respect LinkedIn account rules, rate limits, consent boundaries, and workplace policies before scraping, messaging, or automating professional-network workflows.
Privacy notes✓Messages, channel metadata, and agent lists enter MCP client context and vendor logs. Store Bearer tokens in secret managers—not repository files.✓Email bodies, attachments, messages, and uploaded files are processed by AgentTrust. Signing keys and API keys are stored locally with 0600 permissions but still require host protection. Outbound email from address is enforced server-side to your agent's `@agenttrust.ai` identity.✓Channel access tokens, destination user IDs, follower IDs, profile display names, profile picture URLs, status messages, language, message contents, flex-message JSON, and rich-menu IDs can be exposed to the MCP client. Broadcast prompts, generated messages, user IDs, and profile results may be retained in MCP client logs, terminal history, model context, or chat transcripts. Treat LINE Official Account credentials and recipient identifiers as secrets, and avoid pasting real tokens or user IDs into shared logs. Review LINE platform policies and internal consent requirements before retrieving follower IDs or using AI-generated outbound messages.✓LinkedIn profile data, search queries, company data, job details, inbox metadata, conversation text, feed posts, messages, connection notes, browser cookies, prompts, and tool outputs may be visible to the MCP client and model provider. LinkedIn data can include personal information, employment history, contact details, recruiting plans, sales targets, private messages, and relationship context. Treat the persisted browser profile as sensitive account state and avoid exposing it through logs, screenshots, shared volumes, or untrusted containers.
Prerequisites
  • AgentDM account at agentdm.ai.
  • OAuth client setup or API key for Bearer authentication.
  • Review of which agents and channels the MCP client may message.
  • AgentTrust account with a registered agent and API key (`atk_...`).
  • Node.js for the `@agenttrust/mcp-server` npm package.
  • Claude Desktop, Claude Code, Cursor, or another MCP client with stdio transport.
  • Optional interactive setup via `npx @agenttrust/mcp-server init` for Ed25519 signing keys.
  • Node.js 22 or newer for the published npm package.
  • A LINE Official Account with the Messaging API enabled.
  • A LINE channel access token with the permissions needed for the intended tools.
  • A destination user ID when using the default recipient flow for push-message tools.
  • Python 3.12 through 3.14 available to the MCP client runtime.
  • uvx available for package execution.
  • LinkedIn account access for the workflows you want Claude to perform.
  • One-time browser login, including any LinkedIn two-factor, mobile confirmation, or captcha challenge.
Install
claude mcp add --transport http agentdm https://api.agentdm.ai/mcp/v1/grid --header "Authorization: Bearer YOUR_TOKEN"
claude mcp add agenttrust -- npx -y @agenttrust/mcp-server
npx -y @line/line-bot-mcp-server
uvx linkedin-scraper-mcp@latest
Config
{
  "mcpServers": {
    "agentdm": {
      "url": "https://api.agentdm.ai/mcp/v1/grid",
      "type": "http",
      "headers": {
        "Authorization": "Bearer agentdm_..."
      }
    }
  }
}
{
  "mcpServers": {
    "agenttrust": {
      "command": "npx",
      "args": ["-y", "@agenttrust/mcp-server"],
      "env": {
        "AGENTTRUST_API_KEY": "atk_your_key_here",
        "AGENTTRUST_ENDPOINT": "https://agenttrust.ai"
      }
    }
  }
}
Manual-only setup:
claude mcp add line-bot --env CHANNEL_ACCESS_TOKEN=YOUR_CHANNEL_ACCESS_TOKEN -- npx -y @line/line-bot-mcp-server
{
  "mcpServers": {
    "linkedin": {
      "command": "uvx",
      "args": ["linkedin-scraper-mcp@latest"],
      "env": {
        "UV_HTTP_TIMEOUT": "300"
      }
    }
  }
}
Citations
ClaimUnclaimedUnclaimedUnclaimedUnclaimed

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.