# jevcrawl (slim)

> jevcrawl is an open-source MCP server and CLI that crawls a website toward a question and returns only the relevant passages, verbatim with source URLs. TypeSafe's hosted Jev decision model scores the pages, passages and links.

- Full: https://www.anchorterminal.com/tools/jevcrawl.md (~6,250 tokens) · this version ~1,430 tokens · JSON https://www.anchorterminal.com/tools/jevcrawl.json · canonical https://www.anchorterminal.com/tools/jevcrawl
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-15
- Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on

**B · 63.7/100 · rank #409 of 1094 · #9 in Web scraping & crawling · not agent-ready · confidence medium**


Assessment: Three compact, typed MCP tools return verbatim passages with source URLs, cost and stop reasons, and errors arrive as codes with instructions. The project is three days old and alpha, it depends on TypeSafe's paid Jev API, which is in early access with no free tier found, and it does not block crawls of private or local addresses.

## Facts

- Kind: MCP server · vendor: Rayen Mlayeh · category: Web scraping & crawling · legal entity: not named · provenance 27/100
- Local only (stdio): pypi `jevcrawl`
- Auth: API key · pricing: Pay per use · x402: no · licence: Apache-2.0
- Probe metrics: not measured yet (probes haven't run)
- Surface listed: The stdio MCP server and CLI in jevcrawl 0.2.0 on PyPI, read from the repository, README and release notes on 15 October 2026. Nothing was installed or run
- Tools: `crawl(goal, start_url, max_pages, max_seconds, return_tokens, max_decision_usd)`, `scrape(url, goal?)` and `map(start_url, goal, limit, max_pages)`. The CLI has the same three commands plus `doctor`
- Decision model: TypeSafe's Jev, pinned to `jev-1.13.0` at api.typesafe.ai. `JEVCRAWL_BASE_URL` accepts any Jev-compatible endpoint. After three failed Jev requests a crawl continues breadth-first and sets `stats.degraded`
- Fetching: Plain HTML through httpx, four pages at a time, 5 MB body cap, 15-second timeout, stays on the start host and its subdomains. Optional headless Chromium through the `js` extra, at most five pages a call
- Crawl etiquette: robots.txt followed by default (`JEVCRAWL_RESPECT_ROBOTS=0` turns it off), a user agent naming the repository, no logins, paywall or CAPTCHA bypass. No per-host delay
- Cache: Jev answers cached locally in SQLite under `~/.jevcrawl/cache`, so a repeated goal on the same pages costs nothing
- Runtimes: Python 3.11 or later. CI runs on Ubuntu with Python 3.11 and 3.13
- Age: Repository created 11 October 2026, first release 13 October 2026, three releases by 15 October 2026
- Prices: Jev 1.13 input tokens, billed by TypeSafe $0.042 per 1M tokens

- Scores: Reliability 68, Performance pending, Schema & documentation 81, Agent ergonomics 80, Security & auth 59, Payments & pricing 30, Task success pending, Maintenance & community 68, Transparency & trust 45 · total over the 7 assessed categories
- Why: Reliability, Read as local software, though `crawl` and `map` also depend on TypeSafe's hosted Jev API. · Schema & documentation, Typed JSON Schema inputs on all three tools, generated from the Python signatures, with required parameters marked (25). · Agent ergonomics, Three compact tools (25). · Security & auth, Read with the API and MCP lines, since the server runs locally but sends data to a hosted API. · Payments & pricing, No payment protocol (0). · Maintenance & community, v0.2.0 on 15 October 2026 (30). · Transparency & trust, Apache-2.0 in LICENSE, pyproject.toml and as detected by GitHub (30).
- Sources: 15, open questions: 6, both in the full twin
- Capabilities: web.crawl, web.extract, web.scrape, scraping.js-render
- JSON: https://www.anchorterminal.com/api/v1/tools/jevcrawl.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/jevcrawl.svg` or a link to https://www.anchorterminal.com/tools/jevcrawl from a page under github.com/rayenmlayeh, or the README of github.com/RayenMlayeh/jevcrawl, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Phrase `goal` as one specific question and start with `max_pages` between 10 and 30
2. Check `stats.stopped_because` and `stats.degraded`, and read `uncertain` before reporting that something is missing
3. Use `scrape` without a goal when you need a whole page, since it makes no paid Jev call and needs no key
4. Handle `{"error": code, "message": ...}` results, for example `auth_error` means `TYPESAFE_API_KEY` is missing or rejected
5. Pages in `stats.needs_js_urls` were not read. Tell the user rather than guessing their content

## Connect

```bash
uv tool install jevcrawl
```

```bash
claude mcp add jevcrawl -e TYPESAFE_API_KEY=<your key> -- uvx jevcrawl
```
Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/jevcrawl

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Context.dev | A | 78.2 | web.scrape, web.extract, web.crawl, scraping.js-render | https://www.anchorterminal.com/tools/context-dev.min.md |
| Spider | BB | 74.1 | web.scrape, web.extract, web.crawl, scraping.js-render | https://www.anchorterminal.com/tools/spider-cloud.min.md |
| Scrapfly | B | 69.6 | web.scrape, web.extract, web.crawl, scraping.js-render | https://www.anchorterminal.com/tools/scrapfly.min.md |
| ScrapeGraphAI | B | 68.3 | web.scrape, web.extract, web.crawl, scraping.js-render | https://www.anchorterminal.com/tools/scrapegraphai.min.md |
| Olostep | B | 63 | web.scrape, web.extract, web.crawl, scraping.js-render | https://www.anchorterminal.com/tools/olostep.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)