User repository on GitHub
An agent skill to discover TypeSafe Jev opportunities, design typed questions, and learn from recent community experiments.
- Forks
- 3
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
CommunityJev Agents & MCP
Jevify — altryne
An agent skill from the ThursdAI community that teaches Choice, Noul, and Score batching, scans repos for movable LLM judgments, and ships Python helpers without extra npm deps.
- Stars
- 42
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- altryne
- Last verified
- Sep 2026
About this project
Give your agent a System One capability. Use Jev during everyday agent work, and find what it can improve in your product. Get started · Community examples · Question design · Read the skill. An agent can have access to Jev and still read every document, load every tool result, and repeat every judgment itself. Jevify teaches it to recognize work that a fast, inexpensive decision model can handle, then gives it the tools to act.
Quoted from the project's README. Read the full README.
Used inJev for Software & DevTools
Highlights
- Install with npx skills add altryne/jevify --skill jevify for Claude Code, Codex, Cursor, and others.
- Skill covers question design, shared-state batching, source preservation, and evaluation habits.
- Python helpers need no third-party packages beyond TypeSafe credentials for live calls.
- Independent of TypeSafe; activation depends on the host agent reading SKILL.md.
Watch out
MIT license. Not the Jev model or an official TypeSafe skill; does not auto-intercept tool output on install.
Questions about this entry
Short answers drawn from the linked project or page.
Prompt text plus optional scripts. The README is explicit that skill activation depends on the host agent and that installation does not intercept tool outputs automatically. What you get is SKILL.md with references, plus Python helpers: scan.py reads local text or JSONL, asks Jev for a reading shortlist with source offsets, and leaves the agent to reason from the selected evidence. Whether the agent reaches for it at all is up to the harness.
No, and the two are complementary rather than overlapping. Jevify is an independent ThursdAI community skill that adds agent execution habits, runnable helpers, and a prompt for scanning your own product for judgments that could move to a cheaper model. TypeSafe's skill covers building, question design, composition, and verification. You do not need the other one, and Jevify is not an official TypeSafe product.
Yes, for design and sizing. scan.py --dry-run needs no key and no network, and without a key the agent can still design questions and dry-run request sizing locally. Live calls need a TypeSafe tool or SDK, or Python 3.10+ with outbound HTTPS, plus TYPESAFE_API_KEY in the launching shell or a git-ignored .env. Installing the skill makes no API calls and grants no permission to send data.
They are author-reported, and the README says they were not reproduced there. The examples table covers a short window of community posts and notes that those are people using Jev, not necessarily people using this skill. The verifiable part is the mechanism rather than the numbers: the scan is documented as a first pass, not an exhaustive review, and it deliberately keeps low-ranked and uncertain samples so you can check for missed evidence.
npx skills add altryne/jevify --skill jevify; the CLI asks which agents to install for (Claude Code, Codex, Cursor, and others) and whether to install for the project or globally. Add -g -a claude-code -y to skip prompts, then start a new agent session.
MIT, an independent skill from the ThursdAI community. It is not the Jev model or an official TypeSafe product, and installation does not intercept tool outputs automatically.
Yes: set JEVIFY_DIR to the skill directory and run python3 "$JEVIFY_DIR/scripts/scan.py" report.txt --query "Migration failures and rollback requirements" --dry-run, which needs no key or network. The live scan supports plain UTF-8 text and JSONL records with a text field.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Hacker News: Jevify skill scans a repo for structured LLM calls and rewrites them to Jev providers.
soupz0112 pointsJevify skill – Gets your existing agents running on JevRead the thread on Hacker NewsOther listings of this project
Forks and copies of this project in the directory. This entry is the canonical one.
55GitHub stars
Toolkit that makes a model behind an OpenAI-compatible endpoint or running in-process answer typed choice, score, and yes/no questions from next-token probabilities in Jev API shape.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
fidecastro/jevifyPython
fidecastro/jevifyPython5GitHub stars
Inference engine around Qwen2.5-1.5B-Instruct for classification, yes/no judgements, and rubric scoring with 100% schema-valid JSON, no fine-tuning required.
Jev Tools & CLIs#community#awesomejev#catalog-import
Community
Mintzs/jevifyPython
Mintzs/jevifyPythonMore Jev resources like this
20GitHub stars
Claude Code hooks that narrow a roster in YAML, ask Jev which handler owns a prompt, and log the band before anything is blocked.
Jev Agents & MCP#community#python#coding-agent
Community
AgriciDaniel/gatekeeperPython
AgriciDaniel/gatekeeperPython0GitHub stars
MCP server and Claude Code plugin that files each captured note, link or meeting line with one Jev request: kind, project, action and duplicate questions answered in parallel. Routes to a project only at 0.45 or above, marks a duplicate only at 0.70 with a matching item, and stores text verbatim in append-only JSONL.
Jev Agents & MCP#community#mcp#note-taking
Community
dharun-cohere/jev-sortwellTypeScript
dharun-cohere/jev-sortwellTypeScript110.5kGitHub stars
Multi-agent trading research with optional Jev screening of StockTwits and Reddit posts for company relevance before sentiment analysis runs.
Jev Tools & CLIs#community#python#trading
Community
TauricResearch/TradingAgentsPython
TauricResearch/TradingAgentsPython34.9kGitHub stars
Stencil Labs' coding agent embeds TypeSafe Jev for typed checks, routing, and workflow decisions, with measured Jev tokenizer probes in pi-natives.
Jev Agents & MCP#community#typescript#coding-agent
Community
can1357/oh-my-piTypeScript
can1357/oh-my-piTypeScript32.5kGitHub stars
A Claude Code configuration CLI shipping Jev modules for model routing, skill suggestion, tool guardrails, auto-mode gates, and sandbox hooks.
Jev Agents & MCP#community#typescript#coding-agent
Community
davila7/claude-code-templatesPython
davila7/claude-code-templatesPython523GitHub stars
Classifies and splits PDFs, DOCX, and PPTX with local page text from LiteParse and category decisions from Jev.
Jev Tools & CLIs#community#python#document
Community
jerryjliu/docjevPython
jerryjliu/docjevPythonFrom the Jev community
Posts from builders shipping with Jev right now.