User repository on GitHub
A very WIP tool for jev-ification
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Agents & MCP
Traits
A very WIP tool for jev-ification from Harry Solovay: lessons from coding sessions become approved Markdown traits, and a configurable decision model evaluates code one trait at a time while an MCP server handles group-aware approval.
- Stars
- 40
- Language
- TypeScript
- Reading time
- 3 min
- Published by
- Community
- Author
- harrysolovay
- Last verified
- Oct 2026
- Languages
- TypeScript
- Capabilities
- Open sourceRuns locally
Highlights
- Traits live as readable Markdown in named groups configured in traits.ts; the default project group sits at .agents/traits/project/ under version control.
- create_trait presents destination, path and text via MCP form elicitation; nothing is saved until approval and existing traits are never silently rewritten.
- Findings can be compliant, a violation, not applicable or inconclusive; long inputs are reviewed as overlapping excerpts with 1,000 shared characters.
- traits precommit checks staged gitlinks and published submodule commits for Git groups; it is read-only and needs no model credentials.
- GitHub description calls it a very WIP tool for jev-ification; TypeScript, 40 stars, created 2026-10-06, pushed 2026-10-08, no licence listed.
Quickstart
ts
import type { TraitsConfig } from "traits"
export default {
groups: {
project: { _tag: "Local", description: "Project-specific" }
},
layerDecisionModel
} satisfies TraitsConfigWatch out
README reviewed 2026-10-17; the repo labels itself a very WIP proof of concept and lists no licence. The README text never names Jev or TypeSafe: the jev-ification framing comes from the GitHub description, and the decision-model layer is described as configurable rather than fixed to a particular model.
Questions about this entry
Short answers drawn from the linked project or page.
A tool that turns lessons from everyday coding into durable repository guidance. As you and a coding agent work together, corrections become approved traits: concise rules describing what good code means for the project that the agent consults and extends in later sessions.
The GitHub description calls it a very WIP tool for jev-ification, and the README configures a decision-model layer that evaluates one trait against one change or file at a time. The README text in the digest does not name Jev or TypeSafe, so the connection rests on the description and the decision-model design.
Five steps: implement and learn while using existing guidance; propose and reconcile a concise trait against the corpus; apply the lesson by reviewing relevant code against it; validate changes, including staged changes; and investigate findings, fix violations and rerun the review.
create_trait presents the destination group, path and full text through MCP form elicitation; users can approve, decline, cancel or request revisions, and nothing is saved until approval. Clients without form elicitation receive approval-required, where an explicit request containing the exact proposal counts as fallback approval.
Inputs over 12,000 characters are reviewed as overlapping excerpts with 1,000 characters of shared context, and findings report the excerpt index and total. Judgments needing context outside an excerpt stay inconclusive rather than passing, and an empty or incomplete review is not evidence of compliance.
A Git group is a managed submodule described by url, with optional auto-commit and pin options. traits precommit verifies an initialized submodule, a clean checkout, a staged gitlink matching HEAD, and a commit published on the configured remote branch; it never force-pushes and needs no model credentials.
A very WIP proof of concept by harrysolovay: TypeScript, 40 stars, created 2026-10-06 and pushed 2026-10-08, with no licence listed in the GitHub metadata. Installation is traits install-skill, which places the workflow at .agents/skills/traits/SKILL.md and needs no model credentials.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
X: Jev in front of the HeyGen MCP
X: Playwright CLI + Jev vs Playwright MCP
X: A local MCP server that cuts computer use from 23s to 0.8s. Cartwright built Cortex, a local MCP server where…
X: A side panel that drives any site's WebMCP tools
More Jev resources like this
Reusable project context and specialist workflows for the AI tools you already use: agents, skills, rules, and MCP tools for Claude Code, Codex, Copilot, and Cursor, plus Jev decision support that offloads bounded LLM decisions to save tokens.
Jev Agents & MCP#community#github#agents
Community
jmagly/aiwgTypeScript
jmagly/aiwgTypeScriptSix evidence-oriented tools for the pi and omp coding agents, also usable from any MCP client: navigate unfamiliar code, ask typed questions about repository evidence, review completed changes, and select existing tests.
Jev Agents & MCP#community#catalog-import#logicrw
Community
nomenak/jev-toolsTypeScript
nomenak/jev-toolsTypeScriptRule-based code review for AI coding agents, powered by TypeSafe Jev. Three skills — review a PR, review anything, or loop until the engine reports clean — share one local MCP review engine that never reads your repository; the calling agent supplies the diff and the business context.
Jev Agents & MCP#community#catalog-import#logicrw
Community
thebous/jev-flash-reviewTypeScript
thebous/jev-flash-reviewTypeScriptAn experimental coding agent CLI built on opencode whose v2.0.0 architecture adds a JEV decision layer with five decision gates, shadow mode, and ECE calibration, plus a three-level RSI self-evolution loop and six Pi-framework extensions.
Jev Agents & MCP#community#github#typescript
Community
momozi1996/momo-codeTypeScript
momozi1996/momo-codeTypeScriptA local stdio MCP server from bgivenb that runs whole browser and macOS goals for your agent: your agent supplies the goal, exact values, and completion conditions, and Flick loops observe, decide, and act with TypeSafe Jev. TypeScript core with Playwright and Swift drivers.
Jev Agents & MCP#community#github#mcp
Community
bgivenb/flick-computer-useTypeScript
bgivenb/flick-computer-useTypeScriptMCP server that lets Claude Code, Cursor and other clients drive Chrome over the DevTools Protocol, exposing an accessibility-tree menu of stable refs. Its example Jev loop has Jev pick browser actions from that menu, and the README reports about a third fewer tokens and a quarter fewer tool calls in a blind benchmark.
Jev Agents & MCP#community#mcp#browser-automation
Community
Silbercue/public-browserTypeScript
Silbercue/public-browserTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.