> ## Documentation Index
> Fetch the complete documentation index at: https://opencals.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Agents & App Builders

> Agent Skills that teach AI coding tools — Claude Code, Cursor, Codex, Gemini CLI — to build booking integrations on the Opencals API correctly. Install once and let your agent scaffold, wire, and debug.

Most Opencals integrations are now written inside an AI coding agent. **Agent
Skills** give that agent ground-truth knowledge of the Opencals API and SDK — the
real method names, argument shapes, and flows — so it builds working code instead
of guessing.

Skills follow the open [Agent Skills](https://code.claude.com/docs/en/skills)
standard (a folder with a `SKILL.md`), supported by Claude Code, Cursor, OpenAI
Codex, Gemini CLI and more. They live in one public repo:
**[github.com/letsopencals/skills](https://github.com/letsopencals/skills)**.

## The two skills

> **opencals-storefront-api**
>
> Ground truth for any code that talks to the Opencals API / SDK — availability,
> carts, appointments, checkout, discounts, customer accounts, guests &
> attendees, custom durations, reschedule/cancel, rate limits.
> [Details →](https://opencals.com/docs/ai-agents/storefront-api-skill.md)

> **opencals-build-booking-site**
>
> Build a whole booking website — pick an official template, scaffold, wire it
> up, rebrand, deploy. Uses the API skill for exact call shapes.
> [Details →](https://opencals.com/docs/ai-agents/build-booking-site-skill.md)

## The primary journey

**1. Describe your business**

Hand your agent a business idea and some inspiration — screenshots, a
competitor site, a brand feel.

**2. The agent builds it**

`opencals-build-booking-site` picks the closest template and scaffolds it;
`opencals-storefront-api` supplies the exact API calls to wire availability,
booking and checkout.

**3. Rebrand & deploy**

The agent restyles to your inspiration and deploys to Vercel. You add your
Opencals API key. Bookings, checkout and payments work out of the box.

## Install

**Claude Code**

```
/plugin marketplace add letsopencals/skills
/plugin install opencals@opencals
```

Both skills load automatically and trigger on relevant requests.

**Cursor / Codex / Gemini**

Agent Skills are portable — a folder with a `SKILL.md`. Clone the repo and
copy the skill folders into your tool's skills directory:

```bash theme={"theme":{"light":"github-light","dark":"github-dark-default"}}
git clone https://github.com/letsopencals/skills opencals-skills
# copy opencals-storefront-api/ and opencals-build-booking-site/
# into your agent's skills folder
```

> [!NOTE]
>
> The exact skills directory differs per tool — check your tool's current Agent Skills / rules docs for where skill folders go.

**Any agent (manual)**

Clone the repo and point your agent at the relevant `SKILL.md`:

> Follow the skill in `opencals-storefront-api/SKILL.md` when writing Opencals
> API code, and `opencals-build-booking-site/SKILL.md` to scaffold a site.

**v0 / Lovable / Base44**

Each AI app builder has its own setup, with every field ready to copy:

* **[Lovable](https://opencals.com/docs/ai-agents/lovable.md)**: a custom connector with two knowledge files.
* **[v0](https://opencals.com/docs/ai-agents/v0.md)**: attach these skills from skills.sh (`letsopencals/skills`).
* **[Base44](https://opencals.com/docs/ai-agents/base44.md)**: one backend function, an app secret and two workspace skills.

For other builders (Bolt, Replit), paste the
[API reference](https://opencals.com/ai/opencals-storefront-api.md) and
[booking-site playbook](https://opencals.com/ai/lovable-booking-site-playbook.md)
into the project instructions.

## Docs MCP server

The Opencals developer docs are also an MCP server, so an agent can look things
up while it codes instead of relying on what it remembers:

```
https://opencals.com/docs/mcp
```

It speaks streamable HTTP, needs no API key, and exposes two read-only tools:

| Tool | What it does |
| - | - |
| `search_opencals_developer_docs` | Searches every docs page and returns matching passages with links |
| `query_docs_filesystem_opencals_developer_docs` | Reads full pages and the OpenAPI spec with shell-style commands (`cat`, `head`, `grep`) |

**Claude Code**

```bash theme={"theme":{"light":"github-light","dark":"github-dark-default"}}
claude mcp add --transport http opencals-docs https://opencals.com/docs/mcp
```

Add `--scope project` to share it with your team through `.mcp.json`.

**Cursor**

Add this to `.cursor/mcp.json` in your project, or to `~/.cursor/mcp.json`
for every project:

```json theme={"theme":{"light":"github-light","dark":"github-dark-default"}}
{
  "mcpServers": {
    "opencals-docs": {
      "url": "https://opencals.com/docs/mcp"
    }
  }
}
```

**Other clients**

Any MCP client that supports remote (streamable HTTP) servers can connect to
`https://opencals.com/docs/mcp`. Check your client's MCP docs for where the
URL goes.

> [!NOTE]
>
> The MCP server searches documentation only. It can't read or change store
> data or bookings; your code does that through the
> [Storefront API](https://opencals.com/docs/api-reference/introduction.md) with an API key.

Skills and the MCP server work well together: the skills tell the agent how to
build, and the MCP server lets it check a detail it isn't sure of.

## AI app builders

> **[Lovable](https://opencals.com/docs/ai-agents/lovable.md)**
>
> Custom connector + knowledge files

> **[v0](https://opencals.com/docs/ai-agents/v0.md)**
>
> Opencals skills from skills.sh

> **[Base44](https://opencals.com/docs/ai-agents/base44.md)**
>
> Backend function + workspace skills

## For AI agents reading this

The full documentation is available as machine-readable Markdown:

> **[llms.txt](https://opencals.com/docs/llms.txt)**
>
> Structured index of every docs page.

> **[llms-full.txt](https://opencals.com/docs/llms-full.txt)**
>
> The complete docs in one file for maximum context.

The [API reference](https://opencals.com/docs/api-reference/introduction.md) is generated from the OpenAPI
spec — download it via the **Download spec** action, or fetch
[openapi.json](https://opencals.com/docs/api-reference/openapi.json) directly.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.