agent-kit

MCP server · GitHub · ★ 661 · graded 2026-10-27Source linked

Give Claude/Cursor email powers. 27 MCP tools — inbox, send, reply, contacts, search. Free, no signup.

No outside reports on file

Giving AI direct email access is high-risk because it can read private messages, send on your behalf, and be fooled by convincing forgeries or social engineering.

Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee.

Was this verdict for Agent Kit helpful?

What it can access

Local filesNo local folder access notedLOW
Command executionNo shell command runner notedLOW
Network reachConnects to outside service APINOTED
CredentialsRequires API key or tokenREVIEW
Package sourceGitHub (KeyID-AI/agent-kit)PUBLIC

Inferred from this listing’s public package details and description. Check the project’s source before installing.

Install

No npm or PyPI package is listed. Install from source: github.com/KeyID-AI/agent-kit ↗

All setup options ↓

Check this skill Watch it in my stack Compare README badge View project

Everything known

We found nothing about this project in the sources we read (last checked 2026-10-30). That does not mean it has no problems.

0 incidents · 0 advisories · 0 news articles · 0 community threads · checked 2026-10-30

Popularity grade: C (from public signals such as stars and downloads, not a code review). Graded 2026-10-27.

Sources we read: OSV, GitHub Advisory Database, CISA Known Exploited Vulnerabilities, 44 security news feeds and our incident library. Last checked 2026-10-30.

Check a specific version →

About Agent Kit

This MCP server connects Claude or Cursor to your email account, giving AI assistants the ability to read your inbox, send messages, reply to emails, manage contacts, and search messages. It provides 27 different tools for email tasks without requiring you to sign up for a separate service.

Who it’s for

Developers and AI users who want their Claude or Cursor AI assistant to handle email tasks automatically. Best for people comfortable connecting their email directly to AI tools.

Alternatives

Zapier email integrations: Connect email to automated workflows without direct AI access.
Gmail/Outlook native automation: Use official rules and filters instead of third-party AI tools.
Claude's built-in browsing: Ask Claude to help draft emails you send yourself, avoiding direct account access.

Setup

Check these steps against the project’s own instructions before running anything.

Follow the project's README for installation and email account configuration steps, as setup varies by email provider (Gmail, Outlook, etc.).

Why this grade

CC — Moderate risk signals

Signal-based grade (3/7 points): well starred on GitHub, public source code. This reflects adoption and transparency, not a code review.

Popularity and upkeep

31/100

GitHub stars: 661
Listing last refreshed: 2026-10-29

These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.

Risks to watch for

Full email access: The tool can read all your emails, attachments, and contact information if connected to your account.
Sending without review: AI could draft and send emails you didn't explicitly authorize, potentially embarrassing or harmful.
Credential storage: Your email password or authentication tokens are stored somewhere to keep the connection working; if that storage is compromised, your email is exposed.
Phishing and social engineering: An attacker could trick the AI into forwarding sensitive emails or sending money transfer requests by posing as a trusted contact.
No built-in filtering: The tool may not distinguish between genuine and spoofed emails, so scams could trick the AI into taking harmful actions.

How to lower the risk

Least-privilege permissions

Use app-specific passwords instead of your real password if your email provider offers them.
Restrict the tool to read-only mode if you don't need it to send emails.

Network isolation

Run this tool on a trusted device only, not on shared or public computers.
Keep your device's firewall and antivirus up to date.

Monitoring

Regularly check your email account's login history and active sessions.
Set up alerts for any email sent or forwarded by the tool.

Red flags to watch for

Unexpected emails sent from your account that you didn't ask for.
Sudden access to the tool from unfamiliar devices or IP addresses.

Questions

How do I install Agent Kit?

Agent Kit does not publish a standalone npm or PyPI package. Clone or follow the setup instructions in its source repository at github.com/KeyID-AI/agent-kit.

Are there security reports about Agent Kit?

No reports found for this package in our research (last checked 2026-10-30). Agent Kit currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include full email access, sending without review, credential storage.

Is Agent Kit actively maintained?

Agent Kit has 661 GitHub stars. Our directory data was last refreshed on 2026-10-29. Check recent commit activity, release tags, and open issues at github.com/KeyID-AI/agent-kit before pinning it in production.

What are alternatives to Agent Kit?

Notable alternatives to Agent Kit include Zapier email integrations (Connect email to automated workflows without direct AI access), Gmail/Outlook native automation (Use official rules and filters instead of third-party AI tools), Claude's built-in browsing (Ask Claude to help draft emails you send yourself, avoiding direct account access). You can also compare Agent Kit side by side with other MCP servers in the Related tools section on this page.

Maintain or use Agent Kit? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.

GitHub README Markdown:

markdown

[![AgentArmorHQ Risk Grade for Agent Kit](https://agentarmorhq.com/api/badge/keyid-ai-agent-kit)](https://agentarmorhq.com/skills/keyid-ai-agent-kit)

HTML embed:

html

<a href="https://agentarmorhq.com/skills/keyid-ai-agent-kit"><img src="https://agentarmorhq.com/api/badge/keyid-ai-agent-kit" alt="AgentArmorHQ Risk Grade for Agent Kit" height="20" /></a>

Terminal check:

terminal

curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "keyid-ai-agent-kit"

Share with your security lead

Need sign-off to use agent-kit? Copy this note for Slack, a ticket or a pull request.

### AgentArmorHQ Security Review Memo: agent-kit
- **Listing URL:** https://agentarmorhq.com/skills/keyid-ai-agent-kit
- **Type:** App connector
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
- **Public Adoption:** 661 GitHub stars

#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** Requires API key or token
- **Package source:** GitHub (KeyID-AI/agent-kit)

#### Recommended Hardened Setup
- **Source Repo:** https://github.com/KeyID-AI/agent-kit
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.

_Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee._

Key facts

Type: App connector
Publisher: KeyID-AI
Source code: github.com/KeyID-AI/agent-kit
Est. context load: ~1k tokens / turn (~6 tool schemas, 0.5% of 200k context)
Popularity: 661 GitHub stars
Last updated: 2026-10-29
Last graded: 2026-10-27

Community

Discussions for Agent Kit

Ask questions or share configuration tips. Governed by our Community Rules & Terms.

No posts match this filter yet.

Be the first to ask a question or share a working tip.

Was this listing for Agent Kit helpful?

Check before you install, in Claude

Your assistant answers Block, Review or No known issues for any package before it installs it.

Get this check inside Claude