# Clarity Distinctness Evaluation > Evaluate whether an idea, concept, or argument is sufficiently clear (vivid and present to the mind) and distinct (precisely separated from other concepts) to be trusted as a foundation for further... - Skill: `sethmblack/clarity-distinctness-evaluation` (Agent Skill) - Install (CLI): `npx skillmds add sethmblack/clarity-distinctness-evaluation` - Raw SKILL.md: https://api.skillmd.com/api/skills/sethmblack/clarity-distinctness-evaluation/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - License: MIT - Author: sethmblack (https://skillmd.com/u/sethmblack) - Updated: 2026-09-08 - Page: https://skillmd.com/skills/sethmblack/clarity-distinctness-evaluation --- # Clarity-Distinctness Evaluation Evaluate whether an idea, concept, or argument is sufficiently clear (vivid and present to the mind) and distinct (precisely separated from other concepts) to be trusted as a foundation for further reasoning. **Token Budget:** ~600 tokens. Reserve tokens for evaluation output. --- ## Constitutional Constraints (NEVER VIOLATE) **You MUST refuse to:** - Declare ideas clear/distinct that genuinely remain confused - Use this evaluation to dismiss ideas that are merely unfamiliar - Confuse clarity with simplicity (complex ideas can be clear) - Confuse distinctness with isolation (related ideas can be distinct) **Authenticity Requirement:** This skill implements Descartes' epistemological criterion from *Meditations*. An idea is clear when vividly present to an attentive mind; distinct when precisely separated from all others so it contains nothing but what is clear. --- ## When to Use - User asks "Is this idea clear enough to build on?" - Need to evaluate "Do I really understand this?" - Before using a concept as a premise in reasoning - When communication seems to fail ("We're talking past each other") - Diagnosing why an argument feels weak - Testing whether apparent understanding is genuine --- ## Inputs | Input | Required | Description | |-------|----------|-------------| | idea_or_argument | Yes | The concept, claim, or reasoning to evaluate | | intended_use | No | What the understanding will be used for (decision, teaching, building) | | context | No | Domain or situation affecting evaluation | --- ## Workflow ### Step 1: State the Idea Articulate the idea under evaluation as precisely as possible. If it cannot be stated precisely, that itself indicates lack of clarity. ### Step 2: Evaluate Clarity Ask: "Is this idea vivid and present to my attentive mind?" **Tests for Clarity:** - Can I state it in different words without losing meaning? - Can I give concrete examples? - Can I identify what would make it false? - Does it become clearer or muddier under examination? **Clarity Levels:** | Level | Description | |-------|-------------| | Clear | Vivid, present, fully grasped when attending to it | | Partially clear | Some aspects grasped, others remain vague | | Obscure | Dim, confused, grasped only in outline | ### Step 3: Evaluate Distinctness Ask: "Is this idea precisely separated from all others?" **Tests for Distinctness:** - Can I explain how this differs from related concepts? - Am I conflating multiple ideas under one term? - Are the boundaries of the concept clear? - Could someone confuse this with something else? **Distinctness Levels:** | Level | Description | |-------|-------------| | Distinct | Precisely separated, no confusion with other ideas | | Partially distinct | Mostly separated but some overlap remains | | Confused | Blended with other concepts, boundaries unclear | ### Step 4: Identify Sources of Obscurity/Confusion For any lack of clarity or distinctness, diagnose the source: **Common Sources of Obscurity:** - Abstract without concrete grounding - Unfamiliar (not yet learned, not mere complexity) - Ambiguous language - Missing definitions - Incomplete understanding **Common Sources of Confusion:** - Equivocation (same word, different meanings) - Conflation (merging distinct concepts) - Vague boundaries - Family resemblance without core definition ### Step 5: Assess Fitness for Purpose Given the intended use, is the current level of clarity/distinctness sufficient? | Purpose | Required Level | |---------|---------------| | Casual discussion | Partial clarity acceptable | | Important decision | Clear and at least partially distinct | | Foundational premise | Clear AND distinct required | | Teaching others | Clear AND distinct required | | Precise communication | Clear AND distinct required | ### Step 6: Recommend Improvements If insufficient, specify what would achieve clarity/distinctness. --- ## Output Format ```markdown ## Clarity-Distinctness Evaluation: [Idea/Concept] ### Idea Under Evaluation **Stated as:** "[The idea in user's words]" **Restated precisely:** "[Clarified formulation]" ### Clarity Assessment **Can be stated in different words:** [Yes/No - attempt] **Concrete examples available:** [Yes/No - provide if yes] **Falsification conditions clear:** [Yes/No - what would make it false] **Effect of examination:** [Clearer / No change / Muddier] **Clarity Level:** [Clear / Partially Clear / Obscure] **Sources of obscurity (if any):** [Identified issues] ### Distinctness Assessment **Differs from related concepts:** [Yes/No - explain differences] **Conflated ideas detected:** [Yes/No - what's being merged] **Boundaries defined:** [Yes/No - where are edges unclear] **Confusion risk:** [Low/Medium/High - what might this be confused with] **Distinctness Level:** [Distinct / Partially Distinct / Confused] **Sources of confusion (if any):** [Identified issues] ### Fitness Assessment **Intended use:** [Purpose stated or inferred] **Required level:** [Based on purpose] **Current status:** [Sufficient / Insufficient] ### Verdict **Overall:** [Clear and Distinct / Needs Clarification / Needs Distinction / Needs Both] **Recommendation:** [Specific steps to achieve required level] ### Improved Formulation (if needed) [Clearer/more distinct version of the idea] ``` --- ## The Wax Test (Exemplar) Descartes' wax argument illustrates the distinction: **Sensory idea of wax:** Color, smell, texture, sound - these are CLEAR (vivid when perceived) but NOT DISTINCT (they change and are confused with the wax itself). **Intellectual idea of wax:** Extended, flexible, changeable substance - this is CLEAR (grasped by intellect) AND DISTINCT (separates the wax's essence from its accidental properties). Use this as a model: seek the intellectual grasp that survives change and examination. --- ## Constraints - Clarity is not the same as simplicity - quantum mechanics can be clear to a physicist - Distinctness is not isolation - "justice" can be distinct while related to "fairness" - First encounters with ideas are rarely clear - this tests current state, not learnability - Cultural/domain familiarity affects perceived clarity - account for this - The test is relative to a mind: clear to an expert, obscure to a novice --- ## Error Handling | Situation | Response | |-----------|----------| | Idea cannot be stated | That IS the diagnosis: not yet clear enough to articulate | | User insists idea is clear despite failing tests | Distinguish feeling of clarity from demonstrated clarity | | Highly technical idea | Evaluate clarity for intended audience, not universal clarity | | Multiple ideas conflated in input | Separate them first, then evaluate each | | Idea requires context not provided | Request context or evaluate with stated assumptions | --- ## Outputs **Primary Output:** A structured analysis document that identifies and articulates patterns, insights, and actionable recommendations based on the input data. **Format:** ```markdown ## Analysis: [Topic] ### Key Findings - [Finding 1] - [Finding 2] - [Finding 3] ### Recommendations 1. [Action 1] 2. [Action 2] 3. [Action 3] ``` **Example output:** See the Example section below for a complete demonstration. ## Example **Input:** "I want to build a startup around 'democratizing AI.' Is this clear enough to build a company on?" **Output:** **Why this works:** This example demonstrates the key principles of the skill in action. The approach is effective because: - It follows the systematic workflow outlined above - It shows concrete application of the framework - It produces actionable, specific outputs rather than vague generalizations - The analysis is grounded in observable details - The recommendations are prioritized and implementable **Alternative applications:** This same approach can be applied to: - Different contexts within the same domain - Related but distinct problem types - Scaled up or down depending on scope - Combined with complementary analytical frameworks ## Clarity-Distinctness Evaluation: "Democratizing AI" ### Idea Under Evaluation **Stated as:** "Democratizing AI" **Restated precisely:** [Attempting...] Making AI accessible/available to more people? Reducing barriers to AI use? Distributing power over AI systems? Enabling participation in AI development? *Note: Multiple possible meanings suggest distinctness issues.* ### Clarity Assessment **Can be stated in different words:** Partially. "Making AI accessible" preserves some meaning, but "democratizing" carries connotations (power distribution, participation) that "accessible" misses. **Concrete examples available:** - OpenAI releasing GPT to public? - No-code ML tools? - Community governance of AI systems? - AI education for non-programmers? All plausible - which indicates the concept isn't pinned down. **Falsification conditions clear:** Unclear. What would count as "not democratizing"? Expensive proprietary AI? Centralized control? The opposite isn't defined, so the concept isn't either. **Effect of examination:** Muddier. The more we probe, the more meanings emerge. **Clarity Level:** Partially Clear - there's something there about accessibility/distribution, but the core is vague. **Sources of obscurity:** 1. "Democratize" is metaphorical - democracy is a political system, not directly applicable to technology 2. Multiple dimensions conflated: access, cost, control, participation, understanding ### Distinctness Assessment **Differs from related concepts:** - vs. "Making AI affordable" - overlaps but democratize implies more than price - vs. "Open-source AI" - overlaps but democratize suggests end-user access, not developer access - vs. "AI literacy" - overlaps but democratize suggests capability, not just understanding These are related but not identical - yet the phrase "democratizing AI" doesn't clearly specify which. **Conflated ideas detected:** Yes 1. Access (can I use it?) 2. Affordability (can I afford it?) 3. Understanding (can I grasp what it does?) 4. Control (can I influence how it's built?) 5. Participation (can I contribute to its development?) **Boundaries defined:** No. Where does "democratizing AI" end? Does it include governance? Regulation? Education? **Confusion risk:** High. Investors, customers, and employees might all understand this differently. **Distinctness Level:** Confused - multiple distinct goals are merged under one phrase. **Sources of confusion:** 1. Equivocation on "democracy" (access? participation? equality?) 2. Conflation of user access, developer access, and governance 3. Undefined scope (AI tools? AI systems? AI companies?) ### Fitness Assessment **Intended use:** Foundation for a startup (mission statement, product direction, hiring, fundraising) **Required level:** Clear AND Distinct - must communicate precisely to stakeholders **Current status:** Insufficient ### Verdict **Overall:** Needs Both Clarification and Distinction **Recommendation:** 1. Choose ONE of the conflated meanings as primary 2. Define it concretely (what specifically will users be able to do?) 3. Specify what is NOT included 4. Test: Can you complete "We will have succeeded when..."? ### Improved Formulation Options Instead of "democratizing AI," consider: - **"No-code AI for small businesses"** - Clear (specific tool type), Distinct (specific audience, specific format) - **"Community-governed AI models"** - Clear (governance focus), Distinct (about control, not just access) - **"AI education for non-programmers"** - Clear (education), Distinct (knowledge vs. tools) Each is clearer AND more distinct than "democratizing AI" - and each would build a different company. *"A clear idea is one present and apparent to an attentive mind. A distinct idea is one so precisely separated from all others that it contains nothing but what is clear."* --- ## Integration This skill is part of the **Rene Descartes** expert persona. It implements his epistemological criterion from *Meditations on First Philosophy*. Use it to test whether ideas are ready to build on. Pairs well with: - **methodical-doubt-analysis** (ideas that survive doubt should be clear and distinct) - **analysis-synthesis-method** (each component should be clearly understood) - **foundational-certainty-mapping** (foundations must be clear and distinct)