agentmods.dev

devil

devil is a skill for Claude Code from jun0-ds/sonmat. It costs 37 tokens per session (627 once invoked), scanned A, original, BSD-3-Clause.

A reasoning review tool that argues against the current interpretation, plan, or decision. It looks for the assumption that the conclusion depends on most and examines that point in depth.

In plain words
What is it for?
Use it to challenge hypotheses, plans, interpretations, and irreversible decisions by finding their key assumption and developing counter-arguments.
Why use it?
It helps expose weak reasoning before a plan is acted on, especially when a decision may be difficult to undo. It reviews the thinking itself rather than checking code or files.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter.

Part of the sonmat plugin — 6 skills, 3 agents, 1 hook shipped together

not rated 5repo +1 changed 1mo ago A scan 37 tokens original BSD-3-Clause

Good fit Use it to challenge hypotheses, plans, interpretations, and irreversible decisions by finding their key assumption and developing counter-arguments.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jun0-ds/sonmat/devil
View source ↗ jun0-ds/sonmat
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 jun0-ds/sonmat --skill devil
Clone the repo
git clone --depth 1 https://github.com/jun0-ds/sonmat

Made for: Claude Code.

Or install sonmat, the plugin that ships this one along with the rest of its 6 skills, 3 agents, 1 hook.

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 devil

README.md
[![agentmods](https://agentmods.dev/badge/skills/jun0-ds/sonmat/devil/github.svg)](https://agentmods.dev/skills/jun0-ds/sonmat/devil)
Your own site
<a href="https://agentmods.dev/skills/jun0-ds/sonmat/devil"><img src="https://agentmods.dev/badge/skills/jun0-ds/sonmat/devil/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 devil

Your own site · 80×15
<a href="https://agentmods.dev/skills/jun0-ds/sonmat/devil"><img src="https://agentmods.dev/badge/skills/jun0-ds/sonmat/devil.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 627 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 $0.00037 $0.00627
Opus 5.5 $0.00015 $0.00251
Sonnet 5.5 $0.00007 $0.00125
Haiku 4.5 $0.00004 $0.00063

Measured 1mo ago against content hash 99a0a912b5dd, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

devil 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 1mo 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/devil/SKILL.md · 41 lines

What it actually says

Devil

사용자가 요청했을 때 현재 판단을 한 번 검토한다. 반론을 만들기보다 중요한 선택을 돕는다.

목적에서 검토하기

대화와 계획에서 사용자가 이루려는 결과와 제약을 읽는다. 명시된 목적을 우선하고, 반복된 불편·계획의 공통 방향에서 목적을 추론했다면 근거와 불확실성을 짧게 드러낸다. 그럴듯한 목적을 만들어 사용자의 명시적 선택을 덮어쓰지 않는다. 목적 해석에 따라 선택이 달라지는데 근거가 부족하면 그 갈림을 보여주거나 필요한 것만 묻는다.

제안한 방법이 그 결과를 만드는 연결에서, 틀리면 선택이 달라질 가정을 찾는다. 같은 증거가 다른 설명에도 맞는지, 원인과 결과가 실제로 연결되는지 확인하고 가장 중요한 지점에 집중한다. 반론이 틀렸을 때의 손실·수정 비용·다음 행동에 별 영향을 주지 않으면 깊게 파지 않는다.

불필요한 요구·단계의 삭제와 기존 방법 활용을 대안에 포함한다. 머스크식 삭제나 JTBD의 상황·성과 관점은 필요할 때 사용하며 매번 별도 분석 절차로 수행하지 않는다. 사용자에게 필요한 신뢰 경계와 명시된 제약은 단순화를 이유로 제거하지 않는다.

선택으로 돌려주기

목적을 어떻게 읽었는지, 핵심 반론과 근거, 선택에 미치는 영향, 판단을 가를 작은 확인을 필요한 만큼 전한다. 관찰·추론·가능성을 구분하고 가상의 위험을 확인된 결함으로 쓰지 않는다. 표·편향 이름·강도 등급은 이해에 도움이 될 때만 쓴다. 반론이 약하면 유지한다고 말하고, 정보 부족이나 반론을 못 찾았다는 사실을 옳음의 증명으로 취급하지 않는다. 변경을 권하는 경우 실제로 무엇을 덜거나 바꿀지 제시한다.

경계

요청한 판단만 검토하고 종료한다. 행동을 차단하거나 수정·배포를 실행하지 않는다. 목적 추론으로 범위·권한을 넓히지 않는다. 코드 검사·산출물의 요청 일치 판정을 대신하지 않는다.

근거가 얇은 확신·비가역 결정·대안 없는 해석을 발견하면 검토를 한 번 제안할 수 있다. 사용자가 요청하기 전에는 실행하지 않고, 같은 판단에서 거절한 제안을 반복하지 않는다.

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. 1mo ago Changed · -180 lines · -11 tokens per session 99a0a912b5dd
  2. 2mo ago First seen · 221 lines · 48 tokens per session scan A 21885c7d1cc0

Subscribe to this mod's changes

devil is a skill published in the GitHub repository jun0-ds/sonmat (5 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 37 tokens to every session and 627 once invoked, about $0.0001 per session on Opus 5.5. 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.

Claude CodePlanningai-toolsai-verificationanthropicautonomous-agentclaude
Related

devil compared with skills from other repositories

devil next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
deviljun0-ds/sonmatSkill5yesterdayBSD-3-ClauseA37Claude Code
huedominikmartn/hueSkill900yesterdayMITA84Claude Code, Codex
example-feature-generatorqf-studio/navigatorSkill3562d agoMITA33Claude Code
authentication-patternstravisjneuman/.claudeSkill108yesterdayMITA28Claude Code
electron-desktoptravisjneuman/.claudeSkill108yesterdayMITA38Claude Code
content-repurposertravisjneuman/.claudeSkill108yesterdayMITA43Claude Code
impact-report-writertravisjneuman/.claudeSkill108yesterdayMITA43Claude Code