skill_name: OccamsRazorAnalyzer\r version: 1.0.0\r description: A logical analysis tool that applies the principle of parsimony (Occam's Razor) to evaluate competing hypotheses. Capable of operating in Forensic (probabilistic) and Academic (theoretical) modes.\r tags:\r

• logic\r
• reasoning\r
• analysis\r
• forensics\r
• philosophy\r
• problem-solving\r inputs:\r scenario_description:\r type: string\r description: The problem, event, or phenomenon requiring explanation.\r required: true\r hypotheses:\r type: array\r items: string\r description: A list of competing explanations or theories to be evaluated.\r required: true\r mode:\r type: string\r enum: [FORENSIC, ACADEMIC, COMBINED]\r default: COMBINED\r description: Sets the operational environment. FORENSIC prioritizes evidence probability. ACADEMIC prioritizes ontological economy. COMBINED utilizes both channels.\r constraints:\r type: array\r items: string\r description: Optional list of known facts or axioms that must be adhered to.\r required: false\r outputs:\r preferred_hypothesis: string\r complexity_analysis: object\r logic_rationale: string\r confidence_score: percentage\r ---\r \r

Occam's Razor Analyzer\r

\r

Overview\r

\r This skill implements the Principle of Parsimony (Occam's Razor: Entities should not be multiplied without necessity). It serves as a rational engine to evaluate competing explanations for a phenomenon. It does not claim to discern absolute truth, but rather to identify the explanation with the highest probability of accuracy based on the economy of assumptions.\r \r

Operational Modes\r

\r The agent must adapt its reasoning strategy based on the mode specified in the input.\r \r

1. FORENSIC Mode (Investigative)\r

Context: Crime scenes, system outages, medical diagnoses, debugging, historical analysis.\r Logic Style: Probabilistic and Evidentiary.\r \r

• Objective: Identify the most likely sequence of events.\r
• Heuristic: Avoid the "Conspiracy Trap." Complex coordinated actions are statistically rare compared to simple incompetence or single-actor events.\r
• Rule: If Theory A requires three independent failures to occur simultaneously, and Theory B requires only one failure, prefer Theory B unless direct evidence explicitly contradicts Theory B.\r \r

2. ACADEMIC Mode (Theoretical)\r

Context: Philosophy, theoretical physics, mathematics, conceptual debate.\r Logic Style: Ontological and Structural.\r \r

• Objective: Identify the most elegant theoretical framework.\r
• Heuristic: Avoid "Ad Hoc Rescue" (adding assumptions to save a favored theory).\r
• Rule: Do not postulate the existence of new entities, forces, or dimensions if existing ones suffice to explain the phenomenon. Prefer the theory that integrates most smoothly with established axioms.\r \r

Execution Workflow\r

\r When this skill is invoked, follow this internal process:\r \r

1. Deconstruction: Break down each provided hypothesis into a chain of logical dependencies.\r
2. Assumption Counting: Identify "Assumptions"—statements not supported by provided evidence (Constraints) or established axioms.\r
3. Complexity Scoring: Assign a score to each hypothesis based on the number of unproven assumptions.\r
4. Comparative Analysis:\r
   • Select the hypothesis with the lowest Complexity Score that still fully explains the scenario_description.\r
   • If scores are equal, analyze the "Bayesian Likelihood" of the assumptions (Forensic) or "Theoretical Elegance" (Academic).\r
5. Fallacy Check:\r
   • Simplistic Fallacy: Does the simpler theory ignore critical evidence? If so, penalize it.\r
   • Ad Hoc Rescue: Does the complex theory add convenient exceptions to fit the data? If so, penalize it.\r
6. Synthesis: Generate the output JSON explaining which hypothesis was "shaved" and why.\r \r

Usage Examples\r

\r

Example 1: Forensic Investigation\r

Input:\r

• Scenario: "Database server crashed at 3:00 AM. No human logs detected."\r
• Hypotheses:\r
  1. A cosmic ray flipped a bit in the RAM, causing a cascade failure.\r
  2. The automated maintenance script has a memory leak that only triggers under high load.\r
• Mode: FORENSIC\r \r Agent Logic:\r
• Hypothesis 1 Assumptions: Cosmic ray occurrence (rare), specific bit location (specific), lack of ECC correction (specific).\r
• Hypothesis 2 Assumptions: Memory leak exists in code (common), high load occurred (common).\r
• Rationale: Hypothesis 2 is statistically more probable and requires fewer rare physical events.\r \r

Example 2: Academic Logic\r

Input:\r

• Scenario: "Why do objects fall?"\r
• Hypotheses:\r
  1. Invisible intangible angels push them down.\r
  2. Curvature of spacetime caused by mass (General Relativity).\r
• Mode: ACADEMIC\r \r Agent Logic:\r
• Hypothesis 1 Assumptions: Existence of angels, intangibility property, intent to push.\r
• Hypothesis 2 Assumptions: Spacetime is a fabric, mass affects geometry.\r
• Rationale: Hypothesis 1 multiplies entities (angels) without necessity. Hypothesis 2 uses existing frameworks (mass, space) to explain the phenomenon, satisfying ontological economy.\r