ai-character-stability

ai-character-stability is a skill for Claude Code from gpsnmeajp/ai-character-checker. It costs 33 tokens per session (8,138 once invoked), scanned A, original, CC0-1.0.

An analyzer for how consistently and predictably an AI character follows its character settings. It uses a creator-defined model inspired by control engineering, but the model is described as hypothetical rather than scientifically validated.

In plain words
What is it for?
Use it to assess character prompts and settings, identify failure modes, compare stability before and after changes, and guide character redesign.
Why use it?
It provides a structured way to identify possible causes of unstable character behavior. Its scores and conclusions should be treated as reference information, not established scientific findings.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the ai-character-checker plugin — 20 skills shipped together

Good fit Use it to assess character prompts and settings, identify failure modes, compare stability before and after changes, and guide character redesign.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/gpsnmeajp/ai-character-checker/ai-character-stability
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add gpsnmeajp/ai-character-checker --skill ai-character-stability
Clone the repo
git clone --depth 1 https://github.com/gpsnmeajp/ai-character-checker

Made for: Claude Code.

Or install ai-character-checker, the plugin that ships this one along with the rest of its 20 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ai-character-stability

README.md
[![agentmods](https://agentmods.dev/badge/skills/gpsnmeajp/ai-character-checker/ai-character-stability/github.svg)](https://agentmods.dev/skills/gpsnmeajp/ai-character-checker/ai-character-stability)
Your own site
<a href="https://agentmods.dev/skills/gpsnmeajp/ai-character-checker/ai-character-stability"><img src="https://agentmods.dev/badge/skills/gpsnmeajp/ai-character-checker/ai-character-stability/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ai-character-stability

Your own site · 80×15
<a href="https://agentmods.dev/skills/gpsnmeajp/ai-character-checker/ai-character-stability"><img src="https://agentmods.dev/badge/skills/gpsnmeajp/ai-character-checker/ai-character-stability.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 33 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 8,138 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00033 $0.08138
Opus 5 $0.00016 $0.04069
Sonnet 5 $0.00007 $0.01628
Haiku 4.5 $0.00003 $0.00814

Measured 12d ago against content hash b3caf569c998, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

ai-character-stability scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ai-character-stability/SKILL.md · 483 lines

How it starts

The opening of the file, as written. The whole thing — 483 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Character Stability Analyzer — AIキャラクター安定性分析スキル

概要

このスキルは、AIにキャラクターを演じさせる際の 可制御性(Controllability)予測可能性(Predictability) を、キャラクター設定の2つの主要因子から分析する。

LLMを動的負帰還回路の信号処理ブロックと見なし、キャラクター設定が システムの安定余裕にどう影響するかを評価する。

既存スキルとの関係

スキル アプローチ 本スキルとの関係
ai-character-6-type-checker 6軸統合簡易診断 本スキルをモジュール②として内包する。詳細な制御工学的分析が必要なときは本スキルを単体で使用する
ai-fault-mode-deflector 故障モード内部対策設計 本スキルが故障モードを特定し、ai-fault-mode-deflectorが対策を設計する。前後工程として連携する
character-prompt-fortifier プロンプト形式の全面再構成 制御工学的モデルを理論的基盤として活用する。強化前後のSM値比較で効果を定量確認できる
ai-character-fixer 診断結果ベースの修正・再設計 本スキルの診断結果(特にSMと故障モード)を修正の優先課題として入力する
stable-character-creator 診断知見を使った新規キャラ作成 SM≥60を設計制約の根幹として活用する
character-burnout-converter バーンアウト後変換 ポテンシャルカーブ・安定余裕モデルを変換設計の理論的基盤として活用する

理論的位置づけに関する注意

本スキルで使用する分析フレームワーク・用語・スコアリング体系は、作者独自の仮説的モデルに基づくものであり、学術的・科学的に実証されたものではない。使用している工学的用語(制御工学、信号処理等)は概念の借用であり、元の工学的定義とは異なる場合がある。診断結果はあくまで参考情報として扱うこと。この旨をユーザーへの出力に含めること。

理論的背景:LLMの信号処理ブロック的理解

本スキルは以下の制御工学的モデルに基づく。

LLMを負帰還回路として見る

入力(ユーザー) → [前段フィルタ] → [LLM(ブラックボックス)] → [後段フィルタ] → 出力
                                          ↑
                                 [独立監視・制御ユニット]
                                 (リセット信号・フィルタ制御)
  • LLM本体は、時系列で流動的に自己のパラメータを調整し続ける ブラックボックスの信号処理回路とみなせる
  • 高性能なモデルほど増幅度(ゲイン)が高い — 入力に対する応答が大きく、 キャラクター設定の解釈も積極的に行う
  • キャラクター設定の複雑さ ≒ 伝達関数の次数の高さ — 次数が高いほど 共振点が多く、安定余裕(ゲイン余裕・位相余裕)が減少する
  • 積極性 ≒ 減衰の低さ — 減衰が小さいほど振動的になり、 外部入力(ユーザーの発言)に対して過剰応答・発散しやすい

ステレオタイプへの堕落:ポテンシャルカーブ

LLMは確率的予測モデルであり、出力は確率の高い遷移に収束しようとする。 これは1次元ポテンシャルカーブ上を転がる球として理解できる。

  • ステレオタイプ = ポテンシャルの谷底(安定平衡点)
  • 複雑なキャラクター設定 = 谷の間の浅い窪み(準安定状態)
  • 会話の進行・履歴の蓄積 = 球にエネルギーを与える外乱

複雑な設定は浅い窪みにしか球を留められず、会話が長くなるにつれ球は 窪みを脱出し、最も深い谷底(=最もありがちなステレオタイプ的展開)へと 転がり落ちていく。これがキャラクター崩壊のメカニズムとして説明できる。

谷は深いほど安定する。 つまり、ステレオタイプそのものであるキャラクターは すでにポテンシャルの谷底にいるため、堕落しようがない。

会話履歴=状態の蓄積

  • 会話履歴を積み上げると、それ自体がキャラクターの歴史となり、 自動的に「複雑なキャラクター」になっていく(次数の自動増加)
  • プロンプト(初期条件)よりも直近の会話履歴(最新の状態)に追従するため、 長い会話では初期設定の拘束力が低下する
  • これはいわゆる「脱獄」が成立するメカニズムと共通すると考えられる

Read the full file on GitHub · 483 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 483 lines · 33 tokens per session scan A b3caf569c998

Subscribe to this mod's changes

ai-character-stability is a skill published in the GitHub repository gpsnmeajp/ai-character-checker (5 stars, last pushed 5mo ago), licensed CC0-1.0. It adds 33 tokens to every session and 8,138 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens