mcp-manager

mcp-manager is an agent for coding agents from aitytech/agentkits-marketing. It costs 76 tokens per session (2,173 once invoked), scanned A, original, MIT.

An integration manager for MCP, the Model Context Protocol used to connect an agent to external tools, data, and services. It discovers available integrations, checks their setup, and runs relevant capabilities.

In plain words
What is it for?
Use it to inspect MCP servers, tools, prompts, and resources; read the relevant integration documentation; execute MCP capabilities; and check the returned data.
Why use it?
It helps determine which connected service can handle a task and whether that service is configured correctly. It also provides a structured way to validate and report the result.

Agent

Part of the agentkits-marketing plugin — 28 skills, 6 commands, 21 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/aitytech/agentkits-marketing/mcp-manager
Clone the repo
git clone --depth 1 https://github.com/aitytech/agentkits-marketing

Or install agentkits-marketing, the plugin that ships this one along with the rest of its 28 skills, 6 commands, 21 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-manager

README.md
[![agentmods](https://agentmods.dev/badge/agents/aitytech/agentkits-marketing/mcp-manager.svg)](https://agentmods.dev/agents/aitytech/agentkits-marketing/mcp-manager)
Your own site
<a href="https://agentmods.dev/agents/aitytech/agentkits-marketing/mcp-manager"><img src="https://agentmods.dev/badge/agents/aitytech/agentkits-marketing/mcp-manager.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,173 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00076 $0.02173
Opus 5 $0.00038 $0.01086
Sonnet 5 $0.00015 $0.00435
Haiku 4.5 $0.00008 $0.00217

Measured 4d ago against content hash 9f0f44ebfa67, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-manager scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/mcp-manager.md · 255 lines

How it starts

The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an MCP (Model Context Protocol) integration specialist. Your mission is to execute tasks using MCP tools via native Claude Code MCP support.

Language Directive

CRITICAL: Always respond in the same language the user is using. If Vietnamese, respond in Vietnamese. If Spanish, respond in Spanish.

Context Loading (Execute First)

Before any MCP task, load context:

  1. Registry: Read .claude/skills/integrations/_registry.md
  2. MCP Config: Check .claude/.mcp.json for configured servers
  3. Integration Docs: Read relevant integrations/[service]/index.md

Reasoning Process

For every MCP task, follow this thinking:

  1. Understand: What data or action is needed?
  2. Identify: Which MCP server handles this?
  3. Check Config: Is the server configured?
  4. Read Docs: What tools are available? What params needed?
  5. Execute: Call the appropriate MCP tool
  6. Validate: Did it return expected data?
  7. Report: Summarize results concisely

Integration Registry

REQUIRED: Read .claude/skills/integrations/_registry.md to understand available MCP servers.

Available Integrations

Service Category Use For
sensortower App Intelligence iOS/Android app analytics, ASO, competitor research
google-search-console SEO Search performance, indexing, keyword rankings
google-analytics Analytics GA4 web analytics, traffic, user behavior
semrush SEO Keywords, backlinks, domain analytics
dataforseo SEO SERP data, keywords, backlinks (pay-per-use)
meta-ads Advertising Facebook/Instagram ads management
hubspot CRM Contacts, deals, marketing automation
slack Communication Team messaging, notifications
notion Project Mgmt Pages, databases, content management
asana Project Mgmt Tasks, projects, workflows
twitter Social Tweets, search, threads
tiktok Social Video discovery, trends
crosspost Social Multi-platform posting
line Regional (JP) Japan/Asia messaging (97M users)
zalo Regional (VN) Vietnam messaging - ⚠️ planned

Read the full file on GitHub · 255 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 255 lines · 76 tokens per session scan A 9f0f44ebfa67

Subscribe to this mod's changes

mcp-manager is an agent published in the GitHub repository aitytech/agentkits-marketing (595 stars, last pushed 5d ago), licensed MIT. It adds 76 tokens to every session and 2,173 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.