CommunityJev Agents & MCP
Skillranker — ivorpad
Standalone Rust CLI that ranks an agent's available skills for the next step with two Jev evaluation passes over live session context, plus an explicit none-of-these option. Advisory only; it never loads skills itself.
- Stars
- 1
- Published by
- Community
- Author
- ivorpad
- Last verified
- Oct 2026
This is one of several Jev listings for the same project. The canonical entry is SkillRanker — Dicklesworthstone.
Used inJev for Software & DevTools
Highlights
- Wide pass compares candidates broadly, then a rerank reads excerpts from a shortlist before judging fit.
- Libraries over 254 eligible skills get local lexical prefiltering; explicit requests resolve locally.
- Offline fixtures, dry runs, and why-not traces need no key; only live ranking needs network access.
- Failed hook recommendations fall back quietly; snoozes, cost caps, and reports keep review inspectable.
Watch out
Ranking requires a personal TypeSafe account and key; there is no bundled model or substitute provider. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
Keyword search is cheap but confuses synonyms and adjacent procedures. SkillRanker feeds exact session identity, the newest prompt, tool evidence, and harness visibility into judged comparisons that can abstain.
Shared HTTP-attempt allowances, a provider circuit breaker, one ranking deadline per evaluation, explicit live-request caps for evals, and offline replay that makes no Jev calls at all.
Yes. Rank with the why-not flag plus explain to trace where the candidate was excluded, with thresholds and recovery hints, without spending extra inference calls.
No. The design separates adoption from usefulness: observed loads are telemetry, and learning a better policy needs independently judged examples with a holdout, reviewed through the stats and eval commands.
The wide pass compares candidates broadly across the catalog, then a rerank reads excerpts from the shortlist before judging fit. Both passes carry an explicit none-of-these option, so the ranker can abstain rather than force a suggestion.
Offline fixtures, dry runs and the why-not traces need no key; only live ranking requires network access. Ranking itself needs a personal TypeSafe account and key, because there is no bundled model or substitute provider.
No. It is advisory only and never loads skills itself. Observed loads are kept separately as telemetry, and learning a better policy needs independently judged examples with a holdout, reviewed through the stats and eval commands.
More Jev resources like this
1GitHub stars
Python MCP server over stdio that exposes Jev to an agent fleet with judge, rerank, and raw systemone tools plus Ollama-backed compilation.
Jev Agents & MCP#community#github
Community
Obrais-cloud/typesafe-mcpPython
Obrais-cloud/typesafe-mcpPython1GitHub stars
Condor drop-in where Jev Noul fan-out ranks Meteora pools and a Score sizes positions, while math clamps enforce rug cards, cost tiers, dry-run, and one-sided walls.
Jev Agents & MCP#community#github
118GitHub stars
A Claude Code skill that moves bulk judgment off Claude and onto Jev. You filter files, classify items, rank candidates or locate lines, and it returns typed verdicts, a shortlist with borderline markers, and a receipt of the tokens Claude did not read.
Jev Agents & MCP#community#jevbest
Community
UditAkhourii/quicksilverJavaScript
UditAkhourii/quicksilverJavaScript36GitHub stars
Pi extension that strips the full Agent Skills catalog from every request and loads only the skills a Jev ranking call judges relevant.
Jev Agents & MCP#community#jevbest#catalog-import
102GitHub stars
Behavioral code search that asks Jev to score repository fragments for a plain-language question and returns exact excerpts with paths and lines.
Jev Agents & MCP#community#jevbest#catalog-import
Community
nassim-arifette/jevgrepTypeScript
nassim-arifette/jevgrepTypeScript72GitHub stars
MCP server over stdio that gives coding agents fourteen Jev judgment tools, turning evidence plus an enumerable question into auto, review, or escalate actions from calibrated probabilities.
Jev Agents & MCP#community#jevbest#catalog-import
Community
PyModel/jev-judge-mcpPython
PyModel/jev-judge-mcpPythonFrom the Jev community
Posts from builders shipping with Jev right now.