# How many GTM tools here have been bench tested? 0. That is the honest answer

> 0 of 293 tools in this directory have been bench tested. What the two honesty tiers mean, why the number is on the front page, and what it takes to change it.

*Markdown twin of the HTML page at the same path. Same content, no navigation, no styling, no scripts. Links below point at other twins. Site map for machines: [llms.txt](../llms.txt). The whole dataset: [directory.json](../data/directory.json).*

---
[Directory](../index.md) /
[Learn](index.md) / What the data says

**What the data says**

# How many of these GTM tools have actually been tested?

**The short answer**

0. Every one of the 293 entries in this directory is RESEARCHED tier, meaning facts from public sources with URLs and no usage claims. Nobody here has run these tools. That number is printed on the front page rather than hidden, because it is what makes the other tier mean something.

## The two tiers, and only two

- **RESEARCHED.** Facts from public sources with URLs. No usage claims. Nobody has run this tool. All 293 entries in this build are this tier.

- **BENCH-TESTED.** Andrew personally ran it on a stated date. Cannot be bought. There are 0 of them.

There is no third tier, no star rating, no score and no featured field, because a field that exists is a field somebody eventually tries to buy.

## What that means for every list on this site

Every capability list here answers one question: what does the vendor say this tool does, and can an agent reach it. It never answers which tool is better. There are no tool versus tool verdicts anywhere on this site and there will not be any until somebody has run both, because a comparison without a test is an opinion in a lab coat.

## The other places this build is thin, named

- 32 entries carry fewer than two independent source URLs. They are named on the methodology page rather than quietly padded.

- 83 entries have an unknown access gate.

- 263 have no documentation URL recorded.

- 1 entry claims an MCP server with no parseable URL anywhere in the entry.

- github_url, github_stars, github_last_commit and github_archived are null on all 293 entries. No repo health claim is made anywhere.

## What would change the number

Andrew running a tool himself on a stated date, and publishing what happened including the parts that did not work. A vendor can offer access so a bench test becomes possible, that offer is recorded in the entry's notes, and the offer buys a test rather than a verdict. The verdict ships whatever it says.

## Sources

- [The GTM MCP Directory, methodology](../methodology.md) this site
- [The GTM MCP Directory, the counted data](../data.md) this site
- [The GTM MCP Directory, submit a tool](../submit.md) this site

Every number on this page is generated from directory.json at build time and carries the date it was baked: 2026-08-25. Nothing is typed by hand, nothing is rounded, and nothing is estimated. The underlying data is [published in full](../data.md). Where the honest answer is a zero, the zero is printed.

## Related questions

- [What percentage of GTM tools can an AI agent actually reach?](what-percentage-of-gtm-tools-can-an-agent-reach.md)
- [What does agent ready mean for a GTM tool?](what-does-agent-ready-mean.md)
- [Which GTM jobs can no tool do through an official MCP server?](which-gtm-jobs-have-no-official-mcp-server.md)
- [How do I check whether a tool has an MCP server?](how-do-i-check-if-a-tool-has-an-mcp-server.md)

## In the directory

- [Methodology](../methodology.md)
- [The data endpoint](../data.md)