generation-audit SKILL.md skillA

This file is byte-identical to the first copy the registry indexed. Same content hash, so the same checks give both the same grade.

generation-audit from shimo4228/akc-cycle is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add shimo4228/akc-cycle/generation-audit@v2.0

Pins v2.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
shimo4228/akc-cycle
Path
skills/generation-audit/SKILL.md
Versions recorded
5
Latest version
v2.0, committed Nov 1, 2026
First version recorded
Oct 7, 2026
Size
10,841 bytes
Repository stars
3
License
MIT

Audit the harness when a new Claude model takes a role (judge, build or everyday).

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

# generation-audit — harness audit at a model generation change

The procedure for Scaffold Dissolution's (`rules/common/akc-cycle.md`) **third trigger = model generation change**.
One run performs two checks, carrying line fixes through to application and asset-level verdicts through to the stocktakes.

| Check | Compared against | Engine |
|---|---|---|
| runtime cross-check (Phase 1–2) | the system prompt and tool descriptions actually loaded at inference time | this skill |
| dated-pattern scan (Phase 3) | the outdated writing patterns listed in the target model's official guide | `/claude-api prompt-audit` |

Anthropic updates the dated-pattern table and the target model's behavior inside the `/claude-api` skill with
each model release, so they are not copied here — this skill holds only this harness's scope, exclusions, application
conventions, and recording, plus the runtime cross-check that prompt-audit does not have. Asset-level verdicts (Retire / Merge, etc.)
are owned by the stocktake for each asset class (ADR-0022); this skill stands on the side that supplies the evidence.

## Phase 0 — Target model and scope
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add shimo4228/akc-cycle/generation-audit@sha256:805e25386f141d9f

The line at the top of this page pins the label v2.0, 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_reuztnbx7ugzgg6a.svg)](https://markdownregistry.com/a/art_reuztnbx7ugzgg6a)

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

Versions

The registry has recorded 5 versions of this file, between Oct 7, 2026 and Nov 1, 2026, about one every 6 days.

versioncommittedcommitsizeaudit
v2.0 latest2026-11-01 1c68dc8 10,841 BA view · diff
v2.02026-10-10 778526b 10,642 BA view · diff
v2.02026-10-08 3b0d119 10,425 BA view · diff
v2.02026-10-08 7b64054 11,100 BA view · diff
v2.02026-10-07 768f856 11,103 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (10841 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

shimo4228/akc-cycle · 3 stars · license MIT · pushed 2026-11-01 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_reuztnbx7ugzgg6a
GET https://markdownregistry.com/api/v1/resolve?ref=shimo4228/akc-cycle/generation-audit
GET https://markdownregistry.com/api/v1/blob/805e25386f141d9f21d9f21e92a4fb471dd031fba6aa6c641b18cdca29b36c64

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: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from shimo4228/akc-cycle

llms-full.txt llms.txt file
shimo4228/akc-cycle · llms-full.txt
git:20261101.1c68dc8 · audit A · 3 stars
llms.txt llms.txt file
shimo4228/akc-cycle · llms.txt
git:20261101.1c68dc8 · audit A · 3 stars
adr-writer SKILL.md skill
shimo4228/akc-cycle · skills/adr-writer/SKILL.md · Record a design decision as an ADR in the repo's docs/adr/, or decide whether it clears the filing bar or belongs in a…
git:20261101.1c68dc8 · audit A · 3 stars
agent-stocktake SKILL.md skill
shimo4228/akc-cycle · skills/agent-stocktake/SKILL.md · Audit the ~/.claude/agents definitions and give each a verdict.
v1.0 · audit A · 3 stars
author-calibrated-eval SKILL.md skill
shimo4228/akc-cycle · skills/author-calibrated-eval/SKILL.md · Polish LLM-written prose with the author's own reading as ground truth.
git:20261008.3b0d119 · audit A · 3 stars
context-sync SKILL.md skill
shimo4228/akc-cycle · skills/context-sync/SKILL.md · Audit a repo's context documents (CLAUDE.md, ADRs, README, graph.jsonld) for role overlap and staleness against the…
git:20261101.1c68dc8 · audit A · 3 stars
harness-boundary SKILL.md skill
shimo4228/akc-cycle · skills/harness-boundary/SKILL.md · Design lens: which layer a harness mechanism belongs in, and whether it outlives the next model or runtime.
git:20261101.1c68dc8 · audit A · 3 stars
harness-stocktake SKILL.md skill
shimo4228/akc-cycle · skills/harness-stocktake/SKILL.md · Take stock of the whole harness in one report and one approval, then apply on a branch (ハーネスの棚卸し).
git:20261101.1c68dc8 · audit A · 3 stars
hook-stocktake SKILL.md skill
shimo4228/akc-cycle · skills/hook-stocktake/SKILL.md · Audit the harness hooks (hooks/*.sh, scripts/hooks/*) and their settings.json wiring, and give each a verdict (hook…
git:20261101.1c68dc8 · audit B · 3 stars
jev-judgment-design SKILL.md skill
shimo4228/akc-cycle · skills/jev-judgment-design/SKILL.md · Design how a closed LLM judgment (relevance, novelty, evidence strength) moves to TypeSafe Jev. Use when replacing an…
git:20261008.3b0d119 · audit A · 3 stars
learn-eval SKILL.md skill
shimo4228/akc-cycle · skills/learn-eval/SKILL.md · Extract a reusable pattern from this session and route it to where something reads it, or drop it. Use when saving what…
git:20261010.778526b · audit A · 3 stars
llm-as-judge SKILL.md skill
shimo4228/akc-cycle · skills/llm-as-judge/SKILL.md · Design pattern for LLM-as-judge evaluators: binary checks as evidence, one named verdict.
git:20261008.3b0d119 · audit A · 3 stars

Every file in shimo4228/akc-cycle

Other files named generation-audit

generation-audit SKILL.md skill
shimo4228/claude-harness · skills/generation-audit/SKILL.md · Audit the harness when a new Claude model takes a role (judge, build or everyday).
v2.0 · audit A · 4 stars

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