jevbooks

← All projects

omni-scraper-mcp

Trilokx/omni-scraper-mcp

One MCP facade for multi-source scraping, normalization and Jev evaluation.

Evaluation & Benchmarking96%Runner-up: Verification & Guardrails

What it is

Omni Scraper is an integration and orchestration layer that gives agents one local MCP server to collect public and explicitly authorized data, normalize records into a NormalizedDocument schema, run the vendored Last30Days research engine, and discover upstream provider tools dynamically.

How it uses Jev

The README lists evaluate_jev as a central MCP tool and a TypeSafe Jev evaluation pipeline, and shows a CLI run with --jev-preset market_opportunity_scanner. It does not specify which typed questions Jev answers, what state is passed, or how the returned probabilities are consumed in code.

Technique worth stealing

Keep agent context small by exposing a curated MCP tool set plus on-demand provider passthrough.

Try it

uv run python cli.py run --query "AI agents" --sources hackernews,github,arxiv,techmeme,rss --jev-preset market_opportunity_scanner

View on GitHub

judged by Jevjev-1.13.0

Evidence

Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.

  • Jev-centricno0.10
  • Shows a System One patternno0.19
  • Handles uncertaintyno0.04
  • Measuredno0.04
  • Runnableyes0.94
  • Worth recommendingno0.19
  • Model replicano0.09
  • Problem scopescore on a 0–2 scale1.12
  • About Jevyes0.91

Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.