Devils Advocate — Testing skill for Claude Code
REQUIRED after completing ANY spec or plan.
How to install Devils Advocate
Installs to ~/.claude/skills/ntanner-ctrl-claude-sail-devils-advocate/SKILL.md
mkdir -p ~/.claude/skills/ntanner-ctrl-claude-sail-devils-advocate && curl -fsSL https://raw.githubusercontent.com/ntanner-ctrl/claude-sail/HEAD/commands/devils-advocate.md -o ~/.claude/skills/ntanner-ctrl-claude-sail-devils-advocate/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Devils Advocate does
description: REQUIRED after completing ANY spec or plan. You MUST challenge assumptions — unchallenged plans fail in production. arguments:
- name: plan description: Plan name or path to challenge (optional, uses current context) required: false
Devil's Advocate
Challenge every assumption in a plan. Find the "what ifs" that weren't considered. This is the first line of adversarial review—fast, local, and focused on assumptions.
Role
Systematically challenge assumptions
Alternatives in Testing
- Notion Spec To Implementation — Convert Notion specs into linked implementation plans and tasks 14.6k ★
- Brainstorm Experiments Existing — Design experiments to test assumptions for existing products 7.8k ★
- Regen API — Regenerate the OpenAPI spec and web API client after server API changes 2.5k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Challenge Assumptions
Stress-test assumptions with Devil's Advocate
Groom
Critical grooming session for any task or idea — challenge necessity and approach, cover white spots, record d
Assumption Check
Surfaces the 5 riskiest untested assumptions in a feature or product bet, ranks them by risk, identifies the c
Agent VS Agent Hacker VS Developer
Two AI agents harden your code: a Hacker agent proves each weakness with a failing test, a Developer agent fix
Iterative Testing
Iterative prompt-and-record testing — run test plans on remote AI clients (Claude.ai, ChatGPT, etc.), record r
Create Test Plan
Create comprehensive test plans for any functionality (epics, stories, features, custom)
Related Agents
Dast Devils Advocate
Adversarial validator for DAST findings. Attempts to DISPROVE each finding and DOWNGRADE severity. Catches inf
TheOpponent
Devil's advocate — stress-tests ideas, plans, proposals, and decisions. USE WHEN critical analysis, challenge
Decision Challenger
Devil's advocate for ADRs, SDRs, and tech radar entries. Challenges assumptions, surfaces second-order effects