# AI assistants

Connect an AI assistant to your campaign data over MCP.

With MCP, an AI assistant can read your campaigns and content directly, without writing code. Any tool that supports remote MCP servers with custom headers can use these two values:

| Field | Value |
| --- | --- |
| URL | `https://api.wefluence.app/v1/mcp` |
| Header | `Authorization: Bearer wfb_live_xxxxxxxx` |

The tool list is in [MCP server](https://docs.wefluence.app/en/reference/mcp/).

## Claude

In Claude on the web or Desktop:

1. Open Connectors settings, then add a custom connector.
2. Enter the MCP URL above and choose no sign-in.
3. Under Request headers, add `Authorization` with the value `Bearer wfb_live_xxxxxxxx`.

On Team and Enterprise plans, an Owner adds the connector and every member uses the same key. The Free plan is limited to one custom connector.

In Claude Code:

```bash
claude mcp add --transport http wefluence https://api.wefluence.app/v1/mcp \
  --header "Authorization: Bearer $WEFLUENCE_API_KEY"
```
## Cursor

Add this to `.cursor/mcp.json`:

```json
{
  "mcpServers": {
    "wefluence": {
      "url": "https://api.wefluence.app/v1/mcp",
      "headers": { "Authorization": "Bearer ${env:WEFLUENCE_API_KEY}" }
    }
  }
}
```
## VS Code

Add this to `.vscode/mcp.json`:

```json
{
  "servers": {
    "wefluence": {
      "type": "http",
      "url": "https://api.wefluence.app/v1/mcp",
      "headers": { "Authorization": "Bearer ${input:wefluence-key}" }
    }
  }
}
```
## Your own agent

In the Claude API, use the MCP connector with the beta header `anthropic-beta: mcp-client-2025-11-20`:

```json
{
  "mcp_servers": [{
    "type": "url",
    "url": "https://api.wefluence.app/v1/mcp",
    "name": "wefluence",
    "authorization_token": "wfb_live_xxxxxxxx"
  }],
  "tools": [{ "type": "mcp_toolset", "mcp_server_name": "wefluence" }]
}
```
Agents on other models can call the API directly. Give them [openapi.yaml](https://docs.wefluence.app/openapi.yaml) as their tool list.

## Several brands

One key reads one account. If you have several brands, add one connection per key, with different names such as `wefluence-brand-a` and `wefluence-brand-b`.

## Ready-to-copy prompts

Daily report, for an assistant that is already connected:

```text
Use the Wefluence tools. Get all active campaigns, then for each campaign use get_campaign_stats for the approved content count and total views, and get_top_content for the top 10 pieces of content with their links and creator names. State the as_of time, and say so if engagement data is incomplete. A null shares value means not available, not zero.
```
Full report as a file:

```text
Use the Wefluence tools. Find the campaign I mention, create its report with create_campaign_report in xlsx format, and give me the download link. Do not pull all the content into the conversation.
```
Build an integration, for any coding assistant:

```text
Help me connect this system to the Wefluence API. First read https://docs.wefluence.app/en/llms.txt and https://docs.wefluence.app/openapi.yaml. Read the key from the WEFLUENCE_API_KEY environment variable and never write it in code or logs. Pull per campaign, follow next_cursor until has_more is false, and wait for Retry-After on a 429. Use the content id as a unique key so re-pulls do not create duplicates. Treat null as data not available, not zero.
```
## Files for AI

Every docs page has a Markdown version. The **Copy page** button at the top right copies it, and can open it straight in an AI assistant.

| File | Contents |
| --- | --- |
| [llms.txt](https://docs.wefluence.app/en/llms.txt) | A compact map of the docs |
| [llms-full.txt](https://docs.wefluence.app/en/llms-full.txt) | All of the docs in one file |
| [openapi.yaml](https://docs.wefluence.app/openapi.yaml) | The machine-readable API contract |
| [openapi.json](https://docs.wefluence.app/openapi.json) | The same contract as JSON |
