# Clai

> Updated 2026-10-03 · type: tool · category: mcp-agents · status: active · rev 1

Clai is an open-source command-line AI harness for terminal power users, run as a standalone CLI or embedded as a Go package for agent workflows.

- Open source: yes (MIT)
- Self-hostable: yes
- Pricing model: free
- Best for: A terminal-first developer or ops person who wants an AI harness that lives in the shell — context engineering and agent workflows driven from the command line — and that can also be embedded as a Go package inside their own tooling.
- Not for: A non-technical user who wants a GUI or a hosted product; this is a developer command-line tool, and a general-purpose one rather than a GTM-specific workflow.
- Last verified: 2026-10-03

- **Canonical:** https://gtmstacker.com/registry/tool/clai/
- **Source:** [baalimago · GitHub](https://github.com/baalimago/clai)
- **Tags:** mcp-agents, ai-infrastructure, cli, agent-harness, context-engineering, self-hostable
- **Repository:** https://github.com/baalimago/clai

## Is Clai open source?

Yes, Clai is open source under the MIT license.

## How much does Clai cost?

Clai is free to use.

## Can I self-host Clai?

Yes, Clai can be self-hosted (the source is available under the MIT license).

## Alternatives & related

- [oc (only-cli)](https://gtmstacker.com/registry/tool/oc-only-cli/)
- [HarnessRouter](https://gtmstacker.com/registry/tool/harnessrouter/)


---

Clai is an open-source command-line AI harness for terminal power users: context-engineering and agent workflows you drive from the shell, as a standalone CLI or embedded as a Go package. Open source: yes (MIT); self-hostable; free OSS. It has ~145 stars and is established and actively maintained, with ~925 commits since 2024.

## What it does

Clai lives in the terminal. Rather than a chat window, it gives command-line users a way to run AI with deliberate context engineering: assembling the right context and driving agent workflows from the shell, where a developer already works. Its second shape is as a Go package, so the same harness can be embedded inside your own tools rather than only invoked as a CLI. It is MIT-licensed and self-hostable in the sense that you run it yourself and point it at your own model access. Open source: yes (MIT); free OSS. This is a general-purpose developer tool, so what it does for a GTM stack depends on what you build with it.

## Provenance

- MIT per repo (license read from GitHub metadata, not guessed); ~145 stars; ~925 commits; created 2024-03, active (pushed 2026-10-03) (github.com/baalimago/clai, verified 2026-10-03).
- Repo description: "Command line artificial intelligence - Context engineering for terminal powerusers"; usable as a standalone CLI or as a Go package.
- Surfaced in the GTM Stacker studio daily pull (2026-10-03 pass); license/facts independently verified 2026-10-03.
- Capabilities are the project's own description (vendor-claim); not independently tested here.

## Why it matters for a GTM stack

Most GTM automation gets built by someone comfortable in a terminal, and a harness that lives there — with context engineering as a first-class idea and a Go package you can embed — is a practical building block for that person. Clai is not a GTM product; it is plumbing a builder can wire into prospecting, enrichment, or reporting scripts and then reuse inside a larger tool. For a technical RevOps or growth engineer who prefers the command line to a dashboard, it is worth a look. The honest read: reach is modest (~145 stars), and because it is general-purpose, the GTM value is entirely in what you assemble on top of it. If you want a turnkey workflow rather than a harness, this is not it.