# Vivly

> Vivly is a noise-to-signal API for the public web. We read every public conversation about a category — Reddit, X, Hacker News, App Store, Play Store, Product Hunt, press — and return structured signal as JSON. Two consumption modes: as an SDK (pip install, REST, MCP), or as a custom briefing for consumer brand teams.

## What Vivly is

Vivly turns the public web into structured signal. Use it via:

- **SDK** (Python, Node, REST) — pay-as-you-go credits, no subscription. Starter pack $50.
- **MCP server** for Claude and other MCP-compatible hosts — give your AI agent eyes on the public web.
- **Custom briefings** — bespoke category reports for consumer brand teams. Currently in production with the San Antonio Spurs.

## Primary endpoints

- `v.search(query, sources, window)` — clustered themes, sentiment, attributed quotes. 1 credit per call.
- `v.dataset(query, sources, format)` — full posts and comment threads, JSONL output, thread shape preserved. 5 credits per call.
- `v.track(query, sources)` — continuous monitoring on a topic. 10 credits per day active.
- `v.embed(content)` — pre-embedded vectors for retrieval. 0.1 credit per call.

## Sources

| Source | Freshness | Fields |
|---|---|---|
| Reddit | 5 min | post · comments · score · depth · author |
| X | 10 min | tweet · replies · likes · author · topic |
| Hacker News | 5 min | story · comments · points · threading |
| App Store | 1 hr | review · rating · version · country |
| Play Store | 1 hr | review · rating · version · country |
| Product Hunt | 1 hr | launch · maker · comments · upvotes |
| Press | daily | headline · publication · follow-on commentary |

## Pricing

Two tiers, no free tier:

1. **Top-up credits** (self-serve, pay-as-you-go) — Starter $50 / Builder $200 / Scale $500. Credits don't expire.
2. **Custom** (B2B, talk-to-us) — bespoke briefings, partnerships, dedicated sourcing.

## Documentation

- [Home](https://vivly.in/) — what Vivly is, two ways in
- [Developers](https://vivly.in/developers) — SDK quickstart, source coverage matrix, real SDK calls from case studies
- [Pricing](https://vivly.in/pricing) — credit packs, per-call rates, FAQ
- [Case studies](https://vivly.in/case-studies) — published reports built with the SDK

## Case studies

- [Reading the Meta Glasses moment](https://vivly.in/case-studies/meta-glasses) — 50,000 conversations across Reddit, X, App Store, and tech press. Eleven themes clustered. Built with `v.search(query="meta ray-ban quest orion", sources=[…], window="6w")`.
- [Structuring social data for AI](https://vivly.in/case-studies/structuring-social-data-for-ai) — Vivly × Aquin partnership. 1,547 Reddit and Hacker News threads turned into a fine-tuning dataset. Built with `v.dataset(query="meta ray-ban smart glasses", sources=[…], format="jsonl")`.

## Who Vivly is for

- AI agent builders shipping product (research agents, monitoring agents, brand-intelligence agents)
- Quant and hedge fund data engineers (alpha signal from social conversation)
- Fine-tuning and dataset companies (clean, structured social data for training)
- Indie devs building monitoring tools
- Consumer brand category strategists (custom-briefing tier)

## Contact

founders@vivly.in