> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://api.docs.siftsy.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://api.docs.siftsy.com/_mcp/server.

# Quickstart

> Get a Siftsy Analyst API key and make your first requests.

#### Create an API key

In the Siftsy dashboard, go to **Team Settings → API**. You need to be a team admin on a **Powerhouse** or **Enterprise** plan. This key is for the Siftsy Analyst API.

Create a key and copy it right away. For security, Siftsy only shows the full key once. The default scopes are read-only; if this key should also [add content](/guides/core-concepts/adding-content), turn on `posts:write` / `accounts:write` when you create it.

#### Check that it works

Replace `YOUR_API_KEY` with the key you copied.

```bash
curl -sS \
  -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.siftsy.app/analyst/v1/me"
```

A successful response shows which team the key belongs to, what it can access, and its current rate-limit position (`rateLimit`) — a scheduled sync can check this before it starts. See [Rate limits](/guides/core-concepts/rate-limits).

#### See your collections

```bash
curl -sS \
  -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.siftsy.app/analyst/v1/collections"
```

This returns your campaigns and the groups inside them, just as they appear in Siftsy.

#### Get your posts

```bash
curl -sS \
  -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.siftsy.app/analyst/v1/posts?limit=100"
```

Posts come back in pages of up to 100. If there are more, the response includes a `nextCursor` — pass it back as `cursor` to get the next page.

#### Narrow it down

Add filters to get exactly the posts you want — for example, TikTok posts from one creator with a Siftsy score of 7 or higher:

```bash
curl -sS \
  -H "Authorization: Bearer YOUR_API_KEY" \
  "https://api.siftsy.app/analyst/v1/posts?account=@nike&platform=tiktok&min_score=7"
```

You can also filter by post `link`, posted date, and sentiment percentages. See [Filtering posts](/guides/core-concepts/filtering-posts) for the full list.

## Next steps

* Connect the same key to a chat with the [MCP (beta)](/guides/get-started/mcp). That is for an analyst in Cursor or Claude, not for a warehouse load.
* Read [Filtering posts](/guides/core-concepts/filtering-posts) to find posts by link, account, platform, date, score, or sentiment.
* Read [Retrieving comments](/guides/core-concepts/retrieving-comments) to sample the comments behind an analysis (by mix-bar bucket). This is evidence, not a full comment export.
* Read [Post analysis](/guides/core-concepts/post-analysis) for the content + comment analysis object.
* Read [Scores](/guides/core-concepts/scores) to understand what the numbers mean.
* Read [Adding content](/guides/core-concepts/adding-content) to add posts or a public account from a pipeline (needs the write scopes).
* See [BI sync](/guides/core-concepts/bi-sync) for a recommended order when loading everything into a warehouse or dashboard.
* Browse the [API reference](/api-reference/overview/api-reference) for every endpoint and filter.