# Prism Brand Mentions API (https://www.socialcrawl.dev/platforms/prism/brand-mentions)

> Fans out to content_analysis summary, phrase-trends, and search in parallel, then folds them into one envelope with a computed volume/sentiment trajectory (OLS slope + a noise-aware direction) and an optional LLM digest (`include=digest`). Sentiment is NLP-derived (content_analysis), and mention freshness lags the live web by days. The `computed` block carries stable keys + a `methodology_version` so a monitor can diff runs. `legs[]` reports each leg's status, cost, and latency; if one of the three data legs fails the call still returns a degraded result, and a strict-majority failure refunds half.

TL;DR: `GET /v1/prism/brand-mentions` costs 50 credits per call and returns SocialCrawl's unified JSON schema. Single x-api-key auth, 100 free credits on signup.

## Parameters

| Parameter | Required | Description |
| --- | --- | --- |
| `keyword` | yes | The brand or term to track. Wrap in quotes for exact-phrase semantics. |
| `date_from` | yes | Window start (YYYY-MM-DD). Required by the trend leg. |
| `date_to` | no | Window end (YYYY-MM-DD). Defaults to the latest crawl. |
| `date_group` | no | Trend bucket size: day (default), week, or month. (day \| week \| month) |
| `page_type` | no | Optional surface filter: ecommerce, news, blogs, message-boards, or organization. (ecommerce \| news \| blogs \| message-boards \| organization) |
| `include` | no | Set `digest` to add an LLM narrative summary leg. |

## Code example

```bash
curl "https://www.socialcrawl.dev/v1/prism/brand-mentions?keyword=socialcrawl&date_from=2026-06-01" \
  -H "x-api-key: sc_YOUR_API_KEY"
```

## FAQ

### What is the SocialCrawl Brand Mentions API?

GET /v1/prism/brand-mentions takes one keyword and returns mention volume over time, a sentiment split, top source domains, and recent mentions in a single call. It also adds a computed trajectory — the volume slope and direction — that no single content-analysis endpoint produces on its own.

### How is the sentiment in brand-mentions calculated?

Sentiment is NLP-derived from the underlying content-analysis corpus, not from an LLM, and is returned as a positive/negative/neutral split plus a per-bucket timeline. The optional digest narrates those numbers in one paragraph but never invents new figures.

### How much does a brand-mentions call cost?

Each call is a flat 20 credits and runs three content-analysis legs in parallel. New accounts get 100 free credits with no credit card. If a strict majority of the legs fail, half the cost is refunded automatically; if all three fail, the full 20 credits come back.

### Can I get a volume and sentiment trajectory over time?

Yes. The computed.trajectory block returns a volume_slope (ordinary-least-squares mentions per bucket), a noise-aware volume_direction of rising, flat, or falling, and a negative_share_slope. The timeline carries one bucket per day, week, or month over your date window.

### What does include=digest add to the response?

Adding include=digest appends a short, plain-language summary of the volume and sentiment trajectory written by a fast language model. The price stays a flat 20 credits — the digest is included, not metered — and if the model fails the rest of the response is unaffected.

See the full Prism API: https://www.socialcrawl.dev/platforms/prism

## Pricing

- Standard endpoints: 1 credit per call
- Advanced endpoints: 5 credits per call
- Premium endpoints: 10 credits per call
- 100 free credits on signup, no credit card required. Cached responses cost 0 credits. Credit packs never expire.
- Full pricing: https://www.socialcrawl.dev/pricing

## Explore with AI

Questions this API answers, phrased for an AI assistant:

- What API returns brand mention volume and sentiment for a keyword?
- How do I track brand mentions and sentiment over time via API?
- best brand monitoring API with a sentiment trajectory