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

# Developer

> Code hosting, model providers, and web scraping.

Connect code hosts, model providers, scraping services, and developer MCP servers.

| Connector | What it does | Host | Authentication |
| - | - | - | - |
| ![](https://www.google.com/s2/favicons?domain=github.com\&sz=64) **GitHub** | Repos, issues, and pull requests via the GitHub API. | `api.github.com` | `Authorization: Bearer …` |
| ![](https://www.google.com/s2/favicons?domain=gitlab.com\&sz=64) **GitLab** | Projects and pipelines via the GitLab API. | `gitlab.com` | `PRIVATE-TOKEN` header |
| ![](https://www.google.com/s2/favicons?domain=openai.com\&sz=64) **OpenAI** | Chat, embeddings, and more via the OpenAI API. | `api.openai.com` | `Authorization: Bearer …` |
| ![](https://www.google.com/s2/favicons?domain=openrouter.ai\&sz=64) **OpenRouter** | Unified access to many LLMs (GLM, Claude, o3, and more) via the OpenRouter API. | `openrouter.ai` | `Authorization: Bearer …` |
| ![](https://www.google.com/s2/favicons?domain=anthropic.com\&sz=64) **Anthropic** | Claude models via the Anthropic API. Also send an anthropic-version header from your code. | `api.anthropic.com` | `x-api-key` header |
| ![](https://www.google.com/s2/favicons?domain=scrapfly.io\&sz=64) **Scrapfly** | Web scraping API with JS rendering and anti-bot bypass — scrape pages (e.g. Crunchbase) and get HTML/markdown/JSON. Pass the target as `url` and your key as the `key` query param. | `api.scrapfly.io` | API key as the `key` query parameter |
| ![](https://www.google.com/s2/favicons?domain=apify.com\&sz=64) **Apify** | Run scraping/automation Actors and read their datasets via the Apify API — e.g. LinkedIn, Reddit, and company-profile scrapers. | `api.apify.com` | `Authorization: Bearer …` |
| ![](https://www.google.com/s2/favicons?domain=firecrawl.dev\&sz=64) **Firecrawl** | Scrape, crawl, and search the web (e.g. LinkedIn, Reddit) and get clean, LLM-ready markdown/HTML via the Firecrawl API. | `api.firecrawl.dev` | `Authorization: Bearer …` |
| ![](https://www.google.com/s2/favicons?domain=priorlabs.ai\&sz=64) **Prior Labs** (MCP) | Tabular predictions with TabPFN via the Prior Labs MCP server. | `api.priorlabs.ai` | MCP server at `https://api.priorlabs.ai/mcp/server` |

## Connect one

**1. Open the catalog**

Go to [**Settings → Connector Vault**](https://app.corvic.ai/settings/vault) and switch to **Available to connect**. Type the service's name in the search box, or filter by **API**, **OAuth**, **MCP**, or **Storage/Databases**.

**2. Pick the service**

Selecting a card prefills the host and the header the service expects. If the host is a placeholder such as `yourinstance…`, replace it with your own tenant's host.

**3. Paste your credential and save**

Enter the key or token in the format shown, choose whether the connector is for your organization or just for you, and save. Corvic tests the connection and shows its status on the card.

MCP entries open the **Add MCP server** dialog with the server's address filled in. See [MCP servers](https://docs.corvic.ai/connector-vault/mcp-servers.md).

> [!TIP]
>
> Not listed? Choose **Add Custom Connector** to connect [any HTTPS API](https://docs.corvic.ai/connector-vault/api-keys.md), an [OAuth app](https://docs.corvic.ai/connector-vault/oauth.md), an [MCP server](https://docs.corvic.ai/connector-vault/mcp-servers.md), or [storage or a database](https://docs.corvic.ai/connector-vault/databases.md).


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