Home / notque / vexjoy-agent · skills/infrastructure/browser-jev-automation/SKILL.md · GitHub

browser-jev-automation skillB

browser-jev-automation is agent-read markdown (skill) from notque/vexjoy-agent: Jev-driven browser automation: Jev picks operations, programs execute, a text model writes field values only when Jev cannot pick one from the goal..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# Browser Jev Automation

Zero-dependency browser harness. Programs read the DOM and execute actions, Jev makes every judgment call, an LLM writes text only when the goal does not already contain the value. Built in-house; no pip or npm packages.

## Quick start

```bash
python3 scripts/jev-browser-agent.py \
  --url http://127.0.0.1:8000/ \
  --goal "Sign in with username alice, choose country Canada, and submit." \
  --check-text-contains "Welcome alice" --json-compact
```

Requires `TYPESAFE_API_KEY`, Node 22+, and a local Chromium (Playwright cache or `CHROME_PATH`). Output is one JSON object: `status` (`done|blocked|budget|error`), `reason`, `steps`, `requests`, `final_url`, `verify`, `log` (full probability distributions per step), `usage`.

Measured on a local login form: 5 steps, 7 Jev calls, 1.4 s wall, zero LLM tokens, goal verified.

## Architecture: three tiers applied to browser control

| Tier | Role | Cost |
|---|---|---|
| Programs (tier 1) | Snapshot DOM, extract goal candidates, execute actions, deterministic checks, secret scrubbing | CPU only |
…

Read the whole file at its exact version.

How to install

Latest version
mdr add notque/vexjoy-agent/browser-jev-automation@v1.1.0
Exact content
mdr add notque/vexjoy-agent/browser-jev-automation@sha256:46eccc9c3c8e556a

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_oxq3ril64phmoduu.svg)](https://markdownregistry.com/a/art_oxq3ril64phmoduu)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
v1.1.0 latest2026-09-23 decd9fa 14,061 BB view · diff
v1.1.02026-09-19 28d30f9 12,149 BB view · diff
v1.1.02026-09-19 ab51ee7 12,172 BB view

Audit of the latest version

B  16 of 17 checks passed. Deterministic, no model, same answer every run.
  • fail: No link to a raw IP address (matched: http://127.0.0.1:8000)
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (14061 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No script tag

Source

GitHub

notque/vexjoy-agent · 424 stars · license MIT · pushed 2026-09-23 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_oxq3ril64phmoduu
GET https://markdownregistry.com/api/v1/resolve?ref=notque/vexjoy-agent/browser-jev-automation
GET https://markdownregistry.com/api/v1/blob/46eccc9c3c8e556ab61b9bcfab9ac6e92274712de8d128e947933d4c70afd18a

Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

More from notque/vexjoy-agent

CLAUDE.md claude
notque/vexjoy-agent · CLAUDE.md
git:20260921.59e30cc · audit A · 424 stars
assessment skill
notque/vexjoy-agent · skills/analysis/assessment/SKILL.md · Assessment: read-only inspection, codebase overview, value analysis, health checks, ADR consultation, decision…
git:20260919.28d30f9 · audit A · 424 stars
data skill
notque/vexjoy-agent · skills/analysis/data/SKILL.md · Data analysis and reference enrichment.
git:20260919.28d30f9 · audit A · 424 stars
business-ops skill
notque/vexjoy-agent · skills/business/business-ops/SKILL.md · Business operations: strategy, technology, growth, competitive intelligence, support, finance, HR, legal, operations…
git:20260919.28d30f9 · audit A · 424 stars
code-quality skill
notque/vexjoy-agent · skills/code-quality/code-quality/SKILL.md · Code quality: cleanup, linting, formatting, quality gates.
git:20260919.28d30f9 · audit A · 424 stars
comment-quality skill
notque/vexjoy-agent · skills/code-quality/comment-quality/SKILL.md · Review and fix temporal references in code comments.
git:20260919.28d30f9 · audit A · 424 stars
condense skill
notque/vexjoy-agent · skills/code-quality/condense/SKILL.md · Maximize information density: preserve all instructions, remove prose filler.
git:20260919.28d30f9 · audit A · 424 stars
joy-check skill
notque/vexjoy-agent · skills/code-quality/joy-check/SKILL.md · Validate content framing on joy-grievance spectrum.
git:20260920.c4063a3 · audit A · 424 stars
content-engine skill
notque/vexjoy-agent · skills/content/content-engine/SKILL.md · Repurpose source assets into platform-native social content.
git:20260919.28d30f9 · audit A · 424 stars
content skill
notque/vexjoy-agent · skills/content/content/SKILL.md · Content operations: editorial calendar, marketing, publishing, social media management.
git:20260919.28d30f9 · audit A · 424 stars
headlines skill
notque/vexjoy-agent · skills/content/headlines/SKILL.md · Generate headlines, titles, and subject lines: charge, volume, tighten.
git:20260919.28d30f9 · audit A · 424 stars
image-gen skill
notque/vexjoy-agent · skills/content/image-gen/SKILL.md · AI image generation: Gemini and Nano Banana backends; single/series/batch workflows with prompt-to-disk.
git:20260919.28d30f9 · audit A · 424 stars

Every file in notque/vexjoy-agent

Browse by kind, by grade B, or by owner.