Tom Dörr posted a tweet recommending a GitHub project: gtm-agents. It's a GTM automation plugin set built specifically for Claude Code, and it's free, open-source, under the Apache 2.0 license.


Let's start with the problems it outlines:

- Spending 15+ hours a week manually researching prospects and building lead lists

- Spending 10+ hours a week writing content drafts, emails, and social media posts

- Spending 8+ hours a week analyzing data and building reports in spreadsheets

That adds up to 33 hours. If you work 40 hours a week, that leaves you just 7 hours for actual high-priority work.

That number might sound exaggerated, but if you work in sales, marketing, or growth, you probably can't laugh this off—it hits too close to home.

## What Does It Actually Do

The project claims to have 67 plugins, 92 AI agents, 52 business skills, and 20 workflow orchestrators.

It covers sales, marketing, growth, customer success, and revenue operations.

A few examples:

- Sales: Generate 100 qualified leads in 5 minutes (would take 20 hours manually)

- Marketing: Generate a full month of social media content in 30 minutes (would take 30 hours manually)

- Growth: Complete customer segmentation in 15 minutes (would take 2 days writing SQL manually)

Every use case lists the time saved, and it claims the average user saves $25,000 per year.

## How to Use It

Three steps:

1. Install Claude Code (2 minutes)

2. Add the marketplace: `/plugin marketplace add gtmagents/gtm-agents`

3. Install the plugins you need: `/plugin install sales-prospecting`

Then just describe your task in natural language. For example:

```

/campaign-orchestration:launch-campaign "Q1 Product Launch Campaign"

```

It will automatically coordinate 8 agents: Content Strategist → Designer → Copywriter → Email Marketer → Social Media → Paid Media → Data Analyst → Campaign Manager.

## Thoughtful Design Choices

The project follows Anthropic's open Agent Skills standard. That means these skills aren't locked to Claude Code—they can also be used with OpenAI Codex, Cursor, VS Code Copilot, and GitHub Copilot.

Each skill is a self-contained folder that includes a SKILL.md, optional scripts, references, and templates. It uses progressive disclosure: agents only load the name and description on startup, and only pull the full instructions when the skill is actually needed.

This design is really practical—it doesn't waste tokens, and it doesn't slow down response times.

## A Quick Observation

More and more GTM automation tools like this are popping up. The problem they solve is straightforward: offload repetitive, low-value, time-consuming work to AI so people can focus on strategy and relationship building.

But there's one question: will you actually use the saved time for strategic work?

Chances are, you won't.

The reality is that saved time quickly gets filled up with new "urgent but unimportant" tasks: more meetings, more emails, more alignment checks.

So the real prerequisite for this tool to be useful is that you have to intentionally put that saved time to good use. Otherwise, it just helps you finish all the stuff you shouldn't have been doing in the first place faster.

That said, if you are indeed spending hours every week manually building lead lists, writing emails, and putting together Excel reports, this project is worth trying. It's free, open-source, requires no coding, so there's zero cost to test it out.

GitHub link: https://github.com/gtmagents/gtm-agents

It also has a community Slack and an FAQ, so it's great for beginners.