growth-engineer MCP server
Commit Nov 13, 2026Provides an open-source catalog of go-to-market tools, companies, and workflows that AI agents can discover and execute via markdown files.
Also known as GetBrew/growth-engineer
- Stars
- 218
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 13, 2026
- License
- MIT
What is growth-engineer MCP?
growth-engineer provides an open-source catalog of go-to-market tools, companies, and workflows that AI agents can discover and execute via markdown files.
Summary generated from the project's README
About growth-engineer MCP server
Providing an open-source catalog of go-to-market tools and workflows for AI agents is what growth-engineer enables. It allows developers to connect agents like Claude, ChatGPT, Cursor, or Claude Code to a read-only repository of companies, specific functions, and multi-step processes. Instead of forcing an agent to parse human-centric documentation, this server provides structured markdown files that define exactly how to interact with various APIs, CLIs, or MCP servers.
Users can connect via a Streamable HTTP endpoint without needing any sign-in or API keys. The server includes a search tool to locate specific workflows or companies, a get tool to retrieve the underlying markdown files, and a submit_feedback tool for reporting issues. It also features specialized prompts to help agents contribute new data to the catalog. This setup ensures that complex growth plays, such as outbound motions, can be executed by an agent that follows predefined steps and asks for permission before spending money or changing data.
What you can do with it
- 1Execute complex growth playsAn AI agent can run a multi-step outbound motion by following a specific workflow file.
- 2Discover go-to-market toolsDevelopers can use the search tool to find specific companies or functions available in the catalog.
- 3Automate catalog contributionsAgents can use guided prompts to help research and add new companies or workflows to the repository.
Written from the project's README. Every use case is checked against a quote from it.
Automated checks
Ran Nov 13, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablegetbrew/growth-engineer is public and reachablegithub.com/getbrew/growth-engineer ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-13github.com/getbrew/growth-engineer ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubGo-to-market tools and workflows, written as files any agent can run.
An open-source catalog of go-to-market companies, the functions an agent can call on each one, and workflows that chain those functions into a result. Every entry is a markdown file in this repository.
What is this?
Handing a go-to-market tool to an agent means digging through docs written for people: which key to create, which endpoint to call, whether there is an MCP server, what a call costs. And a growth play that works usually lives in someone's notes, where no agent can run it.
growth.engineer writes both down as markdown files an agent can follow:
| Entry | Example key | What the file holds |
|---|---|---|
| Company | apollo |
Who the company is and how an agent reaches it: MCP server, CLI or API, and the credential each one needs. |
| Tool | apollo/enrich-person |
One function an agent can call: the exact MCP tool, CLI command or API endpoint, and the docs page that names it. |
| Workflow | funding-signal-outbound |
Up to ten steps across tools that reach a result: the one motion it serves, the outcome the user gets, the inputs to ask for, how to set up each tool, and the steps. |
Paste a workflow file into Claude, ChatGPT, Cursor or any other agent and it can run the play. The file tells the agent to ask before it sends a message, spends money or changes data.
Facts come from each vendor's own documentation, and every tool links the page that documents its call.
Connect your agent
The site runs an MCP server at https://www.growth.engineer/mcp
(Streamable HTTP, no sign-in). Its search tool finds workflows, tools and
companies, and get returns a file; neither changes anything. A third tool,
submit_feedback, sends a bug report, request or question to the
maintainers. Every workflow is also an MCP prompt that
takes the workflow's inputs as arguments, and three more prompts
(contribute-workflow, contribute-tool, contribute-company) walk an agent
through adding to the catalog.
| Client | How to add it |
|---|---|
| Claude | Settings → Connectors → Add → Add custom connector. Name it growth.engineer, paste the URL and click Add. |
| Claude Code | claude mcp add --transport http growth-engineer --scope user https://www.growth.engineer/mcp |
| ChatGPT | Plugins in the sidebar → Add → Create MCP App. Paste the URL, pick No authentication, accept the warning and click Create. |
| Codex | codex mcp add growth-engineer --url https://www.growth.engineer/mcp |
| Cursor | In ~/.cursor/mcp.json, add "growth-engineer": { "url": "https://www.growth.engineer/mcp" } under mcpServers. |
Or fetch the files
Every file is public. There is no key and no sign-in.
| URL | Returns |
|---|---|
/workflows/<name>.md |
A workflow, ready to paste into an agent |
/tools/<company>/<name>.md |
One tool |
/companies/<handle>.md |
A company and its tools |
/tags/<namespace>/<slug>.md |
Everything with a tag, such as /tags/capability/enrich-contacts.md |
/llms.txt |
The catalog's definitions and an index of every file |
/llms-full.txt |
Every company, tool and workflow file in one document |
/sitemap.xml |
Every page, with the date it last changed |
A company, tool or workflow page also returns its file to a request with
Accept: text/markdown.
# A workflow file
curl https://www.growth.engineer/workflows/funding-signal-outbound.md
# The same file, by content negotiation
curl -H 'Accept: text/markdown' https://www.growth.engineer/workflows/funding-signal-outbound
Contribute
The catalog lives in this repository, and every addition or fix is a pull request.
| To | Add or edit | Field reference | Agent skill |
|---|---|---|---|
| Share a workflow | workflows/<name>.md |
workflows/README.md |
add-workflow |
| Add a company and its tools | companies/<handle>/company.md and tools/<name>.md |
companies/README.md |
research-company |
| Fix a fact | The file that states it | The same READMEs |
Check your work before you open the pull request. CI runs the same check.
pnpm content:check # every problem in the catalog, each with its file
CONTRIBUTING.md covers the few rules worth knowing first.
How it works
There is no backend and no database. At build time, a compiler in
lib/content/ reads every file, validates it against a strict schema and
resolves every reference. One renderer turns each company, tool and workflow
into the file agents fetch, and Next.js prerenders every page from the same
data. Deploying the site publishes the catalog.
companies/ workflows/ tags.yml
│
▼
lib/content/ read, validate, resolve, derive
│
▼
lib/catalog/render-markdown.ts one renderer, golden-tested
│
├─▶ pages, prerendered at build
├─▶ .md files, /llms.txt, /llms-full.txt
└─▶ /mcp, the MCP server
The one runtime store is optional: an Upstash Redis that counts how often each workflow is copied, for its "Uses" and the Popular order. Without it, the counts are hidden. The footer's newsletter sign-up adds subscribers to Brew from a server action, with the Brew key kept on the server; without a key, it is hidden.
Visits are measured with Vercel Web Analytics and Speed Insights, and AI
crawlers and AI-referred visits with Notra. See
docs/setup.md.
Project layout
growth-engineer/
├─ companies/<handle>/ company.md and tools/<name>.md (logos live on the CDN)
├─ workflows/<name>.md one file per workflow
├─ tags.yml the tag vocabulary
├─ app/ Next.js routes: pages, .md files, /mcp, llms.txt, sitemap
├─ components/ layout, catalog lists, detail pages, UI primitives
├─ lib/
│ ├─ content/ the build-time compiler
│ ├─ catalog/ keys, the markdown renderer, search, loaders
│ ├─ mcp/ the MCP server's tools and prompts
│ ├─ seo/ metadata, structured data, llms.txt
│ └─ usage/ the optional copy counter
├─ tests/ Vitest: golden files, the content suite, schema checks
├─ docs/ architecture, data model, file format, setup
└─ .agents/skills/ add-workflow, research-company
Local development
Prerequisites: Node 22+ and pnpm 11 (corepack enable installs the pinned
version).
pnpm install
pnpm dev # http://localhost:3000
Edits to companies/, workflows/ and tags.yml show on the next refresh.
Add .md to a company, tool or workflow URL to see the file an agent gets.
Nothing needs configuring. .env.example lists the optional
variables:
| Variable | Used for |
|---|---|
NEXT_PUBLIC_SITE_URL |
The origin printed in /llms.txt and page metadata. On Vercel it defaults to the deployment's hostname. |
KV_REST_API_URL, KV_REST_API_TOKEN, KV_REST_API_READ_ONLY_TOKEN |
The copy counter. Without them, the counts are hidden. |
GITHUB_TOKEN |
The star count in the header. Without it, GitHub allows 60 unauthenticated requests an hour. |
NOTRA_GEO_TOKEN |
AI-traffic analytics: the proxy reports AI crawlers and AI-referred visits to Notra. Without it, nothing is sent. |
BREW_API_KEY, BREW_BRAND_ID, BREW_WELCOME_TRIGGER_ID |
The footer's newsletter sign-up, sent to Brew, and the trigger whose automation sends the welcome email. Without a key, the sign-up is hidden; without a trigger, no welcome is sent. |
Scripts
| Script | Does |
|---|---|
pnpm dev |
Starts the dev server |
pnpm content:check |
Parses, validates and renders the whole catalog |
pnpm check |
Biome and a fast typecheck |
pnpm test:run |
The unit suite, including the content checks |
pnpm tsc / pnpm lint |
The full typecheck and lint |
pnpm build |
Production build |
pnpm validate |
Lint, typecheck and tests together |
pnpm hygiene |
Docs links, the content tree, unused code, duplicate dependencies |
Heavy commands wait their turn behind a lock, so several worktrees can run
checks without running out of memory. See
docs/maintainers/validation.md.
Deploy
The site runs on Vercel with the settings in vercel.json.
Every page and file is generated at build time, so merging to main publishes
the catalog. docs/setup.md has the details.
Docs
AGENTS.md: the rules for changing the site, for people and coding agentsdocs/architecture.md: the request path and what prerendersdocs/data-model.md: what the build derives and every rule it enforcesdocs/markdown-files.md: the format of a rendered filedocs/vision.md: why the catalog exists and where it is going
License
MIT © Brew
growth-engineer MCP questions
How do I connect my agent to growth.engineer?
You can connect via Claude, Claude Code, ChatGPT, Codex, or Cursor by using the provided MCP server URL.
Does the MCP server allow making changes to the catalog?
No, the MCP server is read-only and its tools only find or return files without changing anything.
Can I use the workflows without an MCP server?
Yes, every file is public and can be fetched directly via URL or pasted into an agent.