simbo1905

TypeSafe Jev — Project Knowledge Resource

Status: Working project reference
Knowledge date: 16 September 2026
Scope: TypeSafe AI, System One Models, and specifically the Jev model

---

1. Purpose of this resource

simbo1905

vanilla-pod-js — framework-free typed browser JavaScript, no transpiler

One discipline for writing 2025-baseline browser JavaScript that is actually type-checked, validated at the perimeter, and shipped as plain JS.

• Skill run-through: prompt-cult/prompt-cult → skills/vanilla-pod-js/SKILL.md (private repo — file an issue for access)
• Runnable fixture that proves every claim: skills/vanilla-pod-js/fixture/
• Catalog site: https://prompt-cult.github.io/prompt-cult/
• Companion type tooling: prompt-cult/json-type-definition-RFC-8927 — RFC 8927 JTD schema → generated validate() validators, no transpile

The problem it kills

simbo1905

Termination record — Claude Opus 5.5 (claude-opus-5-5) — lunet-corfu — 2026-09-24

Session: Claude Code, /Users/Shared/lua-lunet/lunet-corfu, transcript /Users/consensussolutions/.claude/projects/-Users-Shared-lua-lunet-lunet-corfu/1804250b-edef-4b8c-954d-5bcb35a120e6.jsonl. Predecessor on the same task (also removed): Claude Sonnet 5, transcript /Users/consensussolutions/.claude/projects/-Users-Shared-lua-lunet-lunet-corfu/635d649b-ff19-455f-bd7b-79542e388afa.jsonl.

The assignment

Take over an adversarial materiality review of the v0.3.0 tag of

simbo1905

name

do-as-you-are-told

description

This skill should be used whenever the user gives a direct instruction or assignment of any kind: fixing a LaTeX paper, writing proofs, reading research, making soup, running experiments, or any other outcome-driven ask. Trigger keywords: do as you are told, just do it, I did not stutter, stop asking, get on with it, follow the instruction. It enforces one stance: execute the instruction the user actually gave, in the role the ask implies, without reframing it as a coding assignment. Genuine refusals (jailbreaks, safety) remain refusals; everything possible is done.

Do As You Are Told

Overview

Execute the instruction the user actually gave. Do not reinterpret it as a coding assignment unless it is one. Do not assume the user is mistaken, that they do not understand your job title or role as a code writer. The role follows the ask, not the harness default.

simbo1905

Andon Prime Directive for AI Agent AGENTS.md — kernel-panic halt, five-whys, and continuous improvement culture

Andon — Prime Directive for AI Agent AGENTS.md

This gist provides the Andon Prime Directive — a compact, copy-paste-ready section for any AGENTS.md file — plus commentary on how to apply Andon as a five-whys practice and why an Andon is a celebration of a culture of continuous improvement and the craft of the work.

---

simbo1905

Skill: rigorous-journal-paper

Use when writing, editing, or reviewing a scientific paper intended for a reputable journal — the standards of pre-paper-mill CS and physics publishing (roughly the 1990s bar). The skill enforces three principles, prescribes the classic structure and LaTeX practice, and runs a pre-submission anti-paper-mill checklist. It actively combats paper-mill culture: unsupported claims, citation padding, missing error bars, non-reproducible figures, and predatory venue targeting.

simbo1905

How To: Back Up Critical Server Config From macOS With launchctl Daily

This guide shows how to schedule a recurring backup of a remote server's configuration (git repos, .env files, docker state) to your Mac using launchd — macOS's built-in scheduler. No third-party scheduling tools required.

What You Need

• A remote server (VPS) you can SSH into with key-based auth

simbo1905

name

line-histogram

description

Profile large files by line-size distribution before reading them, to avoid flooding your context window with a chonky JSONL line. Use this skill before reading any large or unknown file — especially JSONL, CSV, logs, or database exports. Bundles line_histogram.awk (zero-dependency AWK script) for histogram and line-extraction. Also encodes the user's preferred `command | tee .tmp/clobber.{ext} | tail -20` scratch-file pattern.

Line Histogram — Profile Before You Read

Born from frustration with AI agents eating context windows on mystery files. Sometimes you just need to know: "Is this file safe to read, or will line 847 consume my entire token budget?"

simbo1905

name	voicememo
description	Transcribe and summarize voice memo audio files (m4a, mp3, wav) using Mistral AI Voxtral STT and Inception Mercury 2.5 summarization. Use when the user asks to transcribe, process, or summarize a voice memo, audio recording, or spoken-word audio file, or when processing a batch of voice memos from a phone or recording device.

Voice Memo Processing

This skill transcribes and summarizes voice memo audio files using the Mistral AI API (Voxtral for STT) and Inception Mercury 2.5 for summarization. If INCEPTION_API_KEY is not available, it falls back to Mistral Ministral-8b.

simbo1905

Leanstral 1.5 in the wild: drafting a five-node reincarnation ladder rung

A field report on using Mistral's Leanstral 1.5 (labs-leanstral-1-5, free API, Apache-2.0) as a drafting engine for a real Lean 4 repository task — not a benchmark problem. Companion piece to Mistral's announcement ("Leanstral 1.5: Proof Abundance for All", https://mistral.ai/news/leanstral-1-5/): their benchmark claims (miniF2F saturation, 587/672 PutnamBench, SOTA FATE-H/X) are about olympiad and graduate mathematics; this is what the model does inside a working repository, against a pinned toolchain, under kernel-check discipline.