# SuiteScrape

SuiteScrape is an AI-ready web scraping SaaS for turning public URLs into markdown, JSON, summaries, raw text, stored jobs, Claude-backed answers, exports, URL monitors, and saved page memory.

## Primary URLs

- Website and workbench: https://suitescrape.com/
- API docs: https://suitescrape.com/docs
- Pricing: https://suitescrape.com/pricing
- Status: https://suitescrape.com/status
- OpenAPI contract: https://suitescrape.com/openapi.json
- API base URL: https://api.suitescrape.com
- Agent onboarding skill: https://suitescrape.com/agent-onboarding/SKILL.md
- Full agent context: https://suitescrape.com/llms-full.txt
- Remote MCP endpoint: https://api.suitescrape.com/mcp with `Authorization: Bearer ss_...`
- Local MCP bridge: run `SUITESCRAPE_API_KEY=ss_... npm run mcp` from the SuiteScrape repo.

## Authentication

Create an API key in the SuiteScrape workbench. Send it as:

```http
Authorization: Bearer ss_your_key
```

Never put API keys in source control, logs, screenshots, prompts, or public issue trackers.

## Most Useful API Calls

- `POST /api/agent/scrape` - scrape one URL, optionally ask a question, and save a stored job.
- `POST /api/agent/ask` - ask Claude about supplied content or a stored job.
- `GET /api/agent/jobs` - list stored jobs with filters and pagination.
- `GET /api/agent/jobs/:id/export?format=markdown` - export a completed job.
- `POST /api/agent/map` - discover same-site URLs from sitemap and page links.
- `POST /api/agent/crawl` - queue a bounded same-site crawl.
- `POST /api/agent/monitors` - create a recurring URL monitor.
- `POST /api/mempalace/ingest` - save a completed scrape into page memory.
- `POST /api/mempalace/ask` - ask across saved page memory.
- Remote MCP: connect a Streamable HTTP MCP client to `https://api.suitescrape.com/mcp` with a SuiteScrape API key bearer token.
- Local MCP tools: `get_plans`, `scrape_url`, `ask_job`, `map_site`, `crawl_site`, `search_jobs`, `export_job`, `create_monitor`.

## Minimal Example

```bash
curl -sS https://api.suitescrape.com/api/agent/scrape \
  -H "Authorization: Bearer ss_your_key" \
  -H "Content-Type: application/json" \
  -d '{"url":"example.com","question":"Return the key facts as bullets.","format":"markdown"}'
```

## Operational Notes For Agents

- Inputs accept full URLs or bare domains such as `example.com`.
- Private and internal URLs are blocked by SSRF validation.
- Normal extraction is fast; Claude is called only when a question is supplied.
- Formats are `markdown`, `json`, `summary`, and `raw`.
- Free plan limits are 50 scrapes and 25 AI queries per month.
- Starter is $9/mo for 1,000 scrapes and 250 AI queries.
- Pro is $29/mo for 10,000 scrapes and 2,500 AI queries.
- Team is $79/mo for 50,000 scrapes and 10,000 AI queries.

Prefer the OpenAPI contract for complete request and response schemas.