| name | critical-thinking-workflow |
| description | Use when stress-testing a high-stakes decision, strategic plan, or argument end-to-end. Triggers on: "critical thinking full process", "decision stress-test", "์ ๋ต ๊ฒํ ์ ์ฒด", ๋
ผ๋ฆฌ ์ ์ฒด ์ ๊ฒ", "high-stakes decision review", "๊ฒฐ์ ์ ๋๋ก ๊ฒํ ํด์ค". |
| type | workflow |
| theme | critical-thinking |
| scenarios | ["Run a full critical thinking review on this decision before I commit","Stress-test this strategic plan โ I want every flaw surfaced","I'm about to make a high-stakes argument โ check it end-to-end","์ด ๊ฒฐ์ ์ ๋๋ก ์ ์ฒด ๊ฒํ ํด์ค, ๋์น ๊ฒ ์๋์ง ํ์ธํ๊ณ ์ถ์ด","์ ๋ต ๊ณํ ์ต์ข
์ ๊ฒ โ ๋
ผ๋ฆฌยทํธํฅยท๊ฒฐ๊ณผ๊น์ง ๋ค ๋ด์ค","์ค์ํ ๋
ผ์ ์ค๋น ์ค์ธ๋ฐ ์ฒ์๋ถํฐ ๋๊น์ง ๋นํ์ ์ผ๋ก ๊ฒํ ํด์ค"] |
| estimated_time | 1-3 hours (full), 15-30 min per step |
| compatibility | {"recommended":["think-tool","mcp-reasoner"],"optional":["sequential-thinking"],"remote_mcp_note":"think-tool์ ๊ฐ ๋จ๊ณ์ ๋ถ์ ํ์ง์ ๋์ด๊ณ , mcp-reasoner๋ ๊ฒฝ์ ์ต์
์ด ์๋ ํ๋จ ๋จ๊ณ(Step 3, Step 6)์์ ํนํ ์ ํจํฉ๋๋ค. Claude ์ค์ โ MCP Servers์์ remote SSE ์๋ํฌ์ธํธ๋ฅผ ์ถ๊ฐํ์ธ์."} |
Critical Thinking Workflow
7-skill end-to-end process: reframe โ surface assumptions โ clean reasoning โ downstream effects โ calibrate confidence โ make tradeoffs explicit.
Workflow Overview
[1] Question Upgrader
|
v
[2] Assumption Extractor
|
v
[3a] Fallacy Detector ----+---- [3b] Bias Auditor
(parallel) | (parallel)
v
[4] Second-Order Thinker
|
v
[5] Epistemic Reasoner
|
v
[6] Tradeoff Articulator
Steps
Step 1 โ Question Upgrader
Skill: question-upgrader
Goal: Verify we are solving the right problem before investing in an answer
Input: Problem as currently framed
Output: Upgraded question(s), named framing weakness, hidden assumption in the question, the question being avoided
Skip if: Framing has already been challenged and a better question is agreed on
"Step 1 ์์" or "์ง๋ฌธ๋ถํฐ ์
๊ทธ๋ ์ด๋ํด์ค"
Step 2 โ Assumption Extractor
Skill: assumption-extractor
Goal: Surface factual, causal, value, and definitional assumptions the argument rests on
Input: Upgraded framing from Step 1
Output: Assumption map across four categories (load-bearing / significant / peripheral); most dangerous assumption named
Skip if: Assumptions have already been explicitly stress-tested
"Step 2 ์์" or "์จ๊ฒจ์ง ์ ์ ์ฐพ์์ค"
Step 3a + 3b โ Fallacy Detector + Bias Auditor (parallel)
Skills: fallacy-detector and bias-auditor
Goal: Check that the argument structure is sound (3a) and the judgment is free of cognitive distortion (3b)
Input: Core argument and rationale from Step 2
Output 3a: Named fallacies with location in argument; narrative fallacy assessment; worst structural error
Output 3b: Active biases across three layers; primary bias and what it distorts
Skip 3a if: The analysis is a plan with no explicit reasoning chains to evaluate
Skip 3b if: No expressed confidence or judgment calls are present
"Step 3 ์์" or "๋
ผ๋ฆฌ ์ค๋ฅ๋ ํธํฅ ๊ฐ์ด ๋ด์ค"
Step 4 โ Second-Order Thinker