Jev SEO
A Rust CLI and MCP server uses Jev to route and score zero-cost SEO and generative-search research tasks.
Added to Jevfast
How it uses Jev
The Rust CLI and MCP server perform deterministic SEO checks and optional Jev evaluations of content/page state. A fan-out request combines a search-intent Choice, a citation-likelihood Score, a direct-answer Noul, and a content-gap Choice; command-specific questions can be merged into that request. Confidence gates can withhold weak judgments, and a hard Jev budget limits spend.
What Jev decides
Illustrative Jev intent questionExample answers · not a recorded Jev response · Source ↗
Question 1 · intent
YOUR APP
INSTRUCTION
Select the primary search intent.
STATE
Caller-supplied SEO evaluation content, after the client prefilters and truncates it; exact fields depend on the command.
JEV · CHOICE
- informational
- commercial
- transactional
- navigational
- insufficient_context
App workflow
- Collect SEO evidence
Commands inspect local content or fetched pages and assemble the relevant page, query, or audit state.
- Build a Jev request
The client prefilters and truncates state, then combines the shared semantic questions with command-specific questions.
- Gate and consume answers
The command reads typed results and applies confidence rules; uncertain judgments are marked for verification or withheld.
- Report or continue
CLI/MCP output presents audit findings, GEO and query evaluations, while the run ledger tracks Jev usage under the configured budget.
Why it is interesting
Jev supplies typed evaluations rather than generated recommendations or prose. The tool can preserve a free local-audit path and make semantic scoring an explicit, budgeted addition.