Dialectic thinking — spawn thesis and antithesis agents to stress-test ideas, then synthesize and verify against code. Use when user says "debate", "argue both sides", "devil's advocate", "stress test this idea", "pros and cons of approach", or wants rigorous evaluation of a design decision.
From dev-toolsnpx claudepluginhub alexei-led/cc-thingz --plugin dev-toolsThis skill is limited to using the following tools:
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Stress-test ideas by spawning competing perspectives, then synthesize and ground-truth against code.
$ARGUMENTS is the question or topic to debate.
If no argument provided, ask the user what they want to evaluate.
<!-- CC-ONLY: begin -->Use TodoWrite to track these 4 phases:
<!-- CC-ONLY: end -->Parse the topic from $ARGUMENTS. Identify:
Frame as a clear binary or spectrum question. Examples:
Spawn two Explore agents in a single message:
<!-- CC-ONLY: begin -->Task(
subagent_type="Explore",
run_in_background=true,
description="Thesis: argue FOR",
prompt="You are arguing FOR: {position A}.
Topic: {framed question}
Build the strongest possible case:
1. Search the codebase for evidence supporting this position
2. Identify concrete benefits with file:line references
3. Anticipate and preempt counterarguments
4. Rate your confidence (high/medium/low) with reasoning
Be specific — cite code, patterns, and constraints you find.
DO NOT hedge or present both sides. Argue your position fully."
)
Task(
subagent_type="Explore",
run_in_background=true,
description="Antithesis: argue AGAINST",
prompt="You are arguing AGAINST: {position A} (i.e., FOR {position B}).
Topic: {framed question}
Build the strongest possible case:
1. Search the codebase for evidence supporting this position
2. Identify concrete risks/costs of the opposing view with file:line references
3. Anticipate and preempt counterarguments
4. Rate your confidence (high/medium/low) with reasoning
Be specific — cite code, patterns, and constraints you find.
DO NOT hedge or present both sides. Argue your position fully."
)
<!-- CC-ONLY: end -->
Collect both results:
<!-- CC-ONLY: begin -->TaskOutput(task_id=<thesis_id>, block=true)
TaskOutput(task_id=<antithesis_id>, block=true)
<!-- CC-ONLY: end -->
Synthesize into a structured verdict:
For any factual claims made by either side (file references, pattern assertions, dependency claims):
DEBATE: {topic}
==============
Thesis: {position A}
Confidence: {high/medium/low}
Key evidence: {1-2 strongest points with file refs}
Antithesis: {position B}
Confidence: {high/medium/low}
Key evidence: {1-2 strongest points with file refs}
Verdict: {position} wins because {reason}
Caveat: {position} would win if {conditions}
Verified claims: {N}/{M} checked, {K} corrected
/debating-ideas Should we split the API into microservices?
/debating-ideas Is it worth adding Redis caching to the auth flow?
/debating-ideas Monorepo vs polyrepo for our frontend packages
If the debate reaches no clear conclusion, present both positions with evidence and let the user decide.
Frame the debate now.