# AI use cases for Hexis (Bevel-Software/Hexis)

Repo: [Bevel-Software/Hexis](https://github.com/Bevel-Software/Hexis) | Use cases: 3 | Page: https://dedupedai.com/use-cases/bevel-software-hexis/

Hexis is a self-hosted, Git-backed control plane that lets a company manage AI agent skills, tool manuals, context and permissions in one place and serve them to any agent over MCP.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Try a shared skill library from your agent in one command: https://dedupedai.com/use-cases/bevel-software-hexis/try-hexis-demo-from-your-agent/
- Self-host a company skill and tool registry: https://dedupedai.com/use-cases/bevel-software-hexis/self-host-hexis-with-docker/
- Let non-engineers propose changes to shared skills: https://dedupedai.com/use-cases/bevel-software-hexis/let-non-engineers-propose-skill-changes/

## How to try a shared skill library from your agent in one command with Hexis

Repo: [Bevel-Software/Hexis](https://github.com/Bevel-Software/Hexis) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README

Install the public Hexis demo MCP server into Cline to see org-managed skills from inside an agent.

**Who it is for:** A platform engineer evaluating ways to distribute skills across the company.

**The problem:** It is hard to judge a skills platform without using it from a real agent.

**The outcome:** A working MCP connection to the demo with browsable plugins.

### Steps

1. Install Cline
2. Run `cline mcp install hexis --transport http https://demo.bevel.software/api/mcp --yes`
3. Ask the agent to list available skills
4. Load one and try it

### Prompt to paste into your agent

```text
Connect to the Hexis demo with `cline mcp install hexis --transport http https://demo.bevel.software/api/mcp --yes`, list the skills it exposes, and load one. Confirm the skill content came from the MCP server.
```

### Caveat

The demo is a shared public instance, so do not put company content into it.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/bevel-software-hexis/try-hexis-demo-from-your-agent/

## How to self-host a company skill and tool registry with Hexis

Repo: [Bevel-Software/Hexis](https://github.com/Bevel-Software/Hexis) | Difficulty: Advanced | Time: 1 hour 30 minutes | Basis: From the README

Deploy Hexis with Docker Compose behind HTTPS so every skill, tool manual and permission lives in your own Git repo.

**Who it is for:** An infrastructure engineer asked to give the company a governed home for AI skills.

**The problem:** Skills shared through chat and copied folders have no review or audit trail.

**The outcome:** A self-hosted control plane where every change has an author, approver and undo path.

### Steps

1. Download docker-compose.yml and .env.example for the release you want
2. Set ADMIN_EMAIL, JWT_SECRET and SECRETS_ENC_KEY, generating secrets with the node crypto command in the README
3. Set DOMAIN and run `docker compose -f docker-compose.yml --profile https up -d`
4. Sign in as admin and connect your agents over MCP

### Prompt to paste into your agent

```text
Deploy Bevel-Software/Hexis on <host> with `docker compose -f docker-compose.yml --profile https up -d`, sign in as admin, and add one skill. Verify the skill change is recorded as a commit in the backing Git repository.
```

### Caveat

An agent can only do what the person running it can do, resolved per file, so permissions must be set before rollout.

Risk, credentials: The .env file holds the admin password and encryption keys, so keep it out of version control.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/bevel-software-hexis/self-host-hexis-with-docker/

## How to let non-engineers propose changes to shared skills with Hexis

Repo: [Bevel-Software/Hexis](https://github.com/Bevel-Software/Hexis) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Give teammates a place to browse plugins and suggest edits that go through review instead of GitHub.

**Who it is for:** An operations lead whose non-technical team wants to improve the prompts their agents use.

**The problem:** Proposing a skill change through GitHub is out of reach for non-technical staff.

**The outcome:** A reviewable suggestion flow with fine-grained access control.

### Steps

1. Deploy Hexis and invite the team
2. Have a teammate browse a plugin and propose a suggestion
3. Review and approve it
4. Confirm agents now load the updated version

### Prompt to paste into your agent

```text
In a Hexis deployment, submit a suggested change to <skill>, approve it as a reviewer, and confirm agents load the new version. Report who changed it and who approved it from the history.
```

### Caveat

This flow needs a running self-hosted deployment; the demo instance is for evaluation only.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/bevel-software-hexis/let-non-engineers-propose-skill-changes/