DESIGN.md@goal-engineer/docsA

DESIGN.md@goal-engineer/docs from kerberosclaw/kc_ai_skills is a DESIGN.md file: the design rules coding agents follow in that repo.

Install it, pinned
mdr add kerberosclaw/kc_ai_skills/goal-engineer/docs/DESIGN.md@git:20261015.be10792

Pins git:20261015.be10792 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 14 of 14 checks passed
Kind
DESIGN.md file
Repository
kerberosclaw/kc_ai_skills
Path
goal-engineer/docs/DESIGN.md
Versions recorded
3
Latest version
git:20261015.be10792, committed Oct 15, 2026
First version recorded
Jun 14, 2026
Size
5,967 bytes
Repository stars
79
License
MIT

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# goal-engineer — 為什麼它不叫 loop-engineer 了

> 現況註記(2026-10-15):本文保留當時的定位與設計脈絡;下文把一般 build 規格/AC 都導往 PRD 的敘述已更新為「repo 功能回 spec,明確需要產品 PRD 才去 prd-create」。現行入口檢核以 [SKILL.md](../SKILL.md) 為準;QA 適用性、align 與收尾主線見[工作流指南](../../workflow-router/docs/workflow.md)。不依本文歷史條目數判斷 Frozen Spec Check 是否完整。

> **English summary:** Design notes for `goal-engineer` (formerly `loop-engineer`). It authors a blind-runnable *dispatch spec* for an unattended, goal-driven evaluator-optimizer loop of the **generate-and-select** kind (produce candidates → grade → iterate by reason-code → human picks the winner). One narrow exception: when a build spec is **already frozen** (approved ADR / locked design / machine-checkable AC) and the only missing piece is the unattended-execution wrapper, it packages a *lean build dispatch* instead of redirecting to a full PRD — gated by the Frozen Spec Check; it still never authors build specs or product decisions. It is the **upstream spec author, not a runtime** — the spec is executed by Claude Code's built-in `/goal`, a headless `claude -p` session, or any unattended agent. Renamed from `loop-engineer` because that name over-claimed the whole "Loop Engineering" idea while…
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add kerberosclaw/kc_ai_skills/goal-engineer/docs/DESIGN.md@sha256:fdc4d5591d6912eb

The line at the top of this page pins the label git:20261015.be10792, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_t2sbmxflx44u7tuz.svg)](https://markdownregistry.com/a/art_t2sbmxflx44u7tuz)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between Jun 14, 2026 and Oct 15, 2026, about one every 62 days.

versioncommittedcommitsizeaudit
git:20261015.be10792 latest2026-10-15 be10792 5,967 BA view · diff
git:20260703.7a7cc052026-07-03 7a7cc05 5,526 BA view · diff
git:20260614.24f1cf42026-06-14 24f1cf4 4,313 BA view

Audit of the latest version

A  14 of 14 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Size between 200 bytes and 200 KB (5967 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 link to a raw IP address
  • pass: No script tag

Source

GitHub

kerberosclaw/kc_ai_skills · 79 stars · license MIT · pushed 2026-10-16 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_t2sbmxflx44u7tuz
GET https://markdownregistry.com/api/v1/resolve?ref=kerberosclaw/kc_ai_skills/goal-engineer/docs/DESIGN.md
GET https://markdownregistry.com/api/v1/blob/fdc4d5591d6912eb6a4c2a83fa063681a7dbd476e53c81b5fbe5b905b97ddbde

Your AI assistant 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.

Guides: what DESIGN.md is and how agents use it · which agent file does what.

More from kerberosclaw/kc_ai_skills

adr SKILL.md skill
kerberosclaw/kc_ai_skills · adr/SKILL.md · Use when a durable technical decision was just made in conversation and should be recorded, or when user asks to…
v0.3.1 · audit A · 79 stars
align SKILL.md skill
kerberosclaw/kc_ai_skills · align/SKILL.md · Reconcile existing repository documentation with the current implementation, especially after verified hotfixes or…
v0.4.0 · audit A · 79 stars
character-lora SKILL.md skill
kerberosclaw/kc_ai_skills · character-lora/SKILL.md · Use when the user wants to build a consistent-identity LoRA for an original character — defining the character…
v0.2.1 · audit A · 79 stars
conference-report SKILL.md skill
kerberosclaw/kc_ai_skills · conference-report/SKILL.md · Use when the user attended one or more sessions of a conference (with audio recordings + slide photos) and needs help…
v0.2.0 · audit A · 79 stars
ctf-kit SKILL.md skill
kerberosclaw/kc_ai_skills · ctf-kit/SKILL.md · Use when the user is solving an authorized CTF / lab reverse-engineering challenge focused on Windows application…
v0.5.0 · audit A · 79 stars
diagnose SKILL.md skill
kerberosclaw/kc_ai_skills · diagnose/SKILL.md · Use when the user reports misbehaving software to investigate — a bug, crash, wrong output, flaky behavior, or…
v0.3.0 · audit A · 79 stars
goal-engineer SKILL.md skill
kerberosclaw/kc_ai_skills · goal-engineer/SKILL.md · Author a self-contained unattended dispatch for a generate-and-select loop (generate candidates, grade, iterate, keep…
v1.3.2 · audit A · 79 stars
gpt-image-gen SKILL.md skill
kerberosclaw/kc_ai_skills · gpt-image-gen/SKILL.md · Use when the user asks to generate OR edit an image via GPT/Codex (e.g. 「叫 gpt 生圖」「幫我用 gpt 生圖」「gpt 畫一個…
v0.6.0 · audit A · 79 stars
grill SKILL.md skill
kerberosclaw/kc_ai_skills · grill/SKILL.md · Use when the user says 先討論 / wants to align on a fuzzy idea before building, or before any non-trivial implementation…
v0.1.0 · audit A · 79 stars
ios-resign SKILL.md skill
kerberosclaw/kc_ai_skills · ios-resign/SKILL.md · Use when the user wants to rebuild, re-sign and reinstall one of their own iOS apps onto their iPhone(s) — typically…
v0.1.1 · audit A · 79 stars
job-scout SKILL.md skill
kerberosclaw/kc_ai_skills · job-scout/SKILL.md · Use when the user is considering applying to, interviewing with, or accepting an offer from a company and wants due…
v1.1.0 · audit A · 79 stars
llm-benchmark SKILL.md skill
kerberosclaw/kc_ai_skills · llm-benchmark/SKILL.md · Use when the user wants to test, compare, or choose local Ollama models for their machine. Checks Ollama/GPU state…
v1.1.0 · audit B · 79 stars

Every file in kerberosclaw/kc_ai_skills

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