could-jev
Claude skill that gives a ten-line gut-check on whether Jev could help with what you are building, citing tested patterns.
This repository is mostly documentation or write-up rather than runnable code.
Overview
could-jev is a Claude skill that answers one question: could Jev help with this? Ask Claude something like "is this a jev thing?" and it returns a short structured verdict with the matching pattern, a sketch of state and questions, the closest prior example, pitfalls, the cheapest test and a bigger idea.
It does not call Jev itself; it is a design aid. The skill bundles a catalogue of 31 Jev patterns (funnels in front of an LLM, censuses instead of samples, locators, paragraph heat maps, pairwise judgments, uncertainty spotlights, agent tripwires) tagged as tested, documented, community or untried, plus a dated capabilities factsheet. You can log your own experiment results so they take priority over generic rules, and a monthly maintenance sweep refreshes the content via pull requests.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Could Jev help with this use case, and which pattern fits?(inferred) | yes · partly · no |
How it works
- Describe your project to Claude and ask if Jev could help
- Skill matches it against its pattern catalogue and factsheet
- Returns verdict, sketch, caveats and cheapest test
- Optionally log your experiment result to personal history
Strengths
- Quick design triage without writing code
- Patterns tagged by evidence level
- Personal history overrides generic advice
Caveats
- Advisory only; does not run Jev
- Factsheet can go stale as Jev changes
Getting started
/plugin marketplace add rlfordon/could-jev then /plugin install could-jev@could-jev in Claude Code.At a glance
- Type
- Agent skill
- Status
- beta
- Language
- Markdown
- Runtime
- Claude Code skill / Claude.ai skill
- Has code
- yes
- Has tests
- no
Use cases and integrations
How curated lists describe it
Claude skill: a quick gut-check on whether Jev (TypeSafe's typed-judgment model) could help with what you're building.
hellogumbo/awesome-jev
Frequently asked questions
Does could-jev call Jev?
No, it is a design gut-check that reasons from a pattern catalogue.
Where does it run?
In Claude Code as a plugin, or in Claude.ai and Claude Desktop as an uploaded skill.
How does it stay current?
A monthly sweep checks TypeSafe docs and community projects and opens a PR with updates.
Related projects
Thirty worked Jev uses in ten levels, from a smart if statement to a pi coding agent that writes its own Jev questions.
ProjectbetaTypeScriptAgent skill that makes a coding agent try Jev first for bounded judgments like classify, rank or gate, and log each attempt.
Agent skillbetaPythonIndependent Markdown reference on the Jev API, patterns and limits, written for engineers and coding assistants to build from.
ServiceAccenture thought piece and site on stack migration with Google Cloud, placing Jev as the System 1 layer that routes, classifies and gates.
MCP serverdiscussionRustCommunity directory of Jev apps, resources and runnable offline examples, with a guide skill and a searchable JevList web UI.
ArticleproductionTypeSafe's official cookbook for a two-stage skill-suggestion workflow that selects a skill and can reject the shortlist.
Articleproduction