analyst

Plugin analyst from marketplace lucid-toolkit in rayk/lucid-toolkit. Commands read from the repo's manifest — 1 month ago.

Skills (1)

• consider

  Selects and applies mental models for structured problem analysis. Triggers when user asks "why", "what if", "how should we", needs systematic problem-solving, or mentions analyzing a situation. MUST BE USED when comparing options, making decisions, or evaluating trade-offs.

Slash commands (5)

• /analyst:assess

  Four-phase rigorous assessment with parallel stress testing

• /analyst:consider

  Analyze problems using parallel multi-agent mental models with voting

• /analyst:debate

  Multi-agent deliberation for complex decisions requiring adversarial exploration

• /analyst:reflect

  Reflect on recent behavior, identify protocol violations, design robust solutions

• /analyst:swarm

  Parallel hypothesis testing for diagnosis problems

Subagents (20)

• model-10-10-10

  Time horizon analysis specialist for decisions with emotional bias. Use when short-term and long-term interests conflict, or emotions are running high.

• model-5-whys

  Root cause analysis specialist using iterative "why" drilling. Use for diagnosis problems, understanding failures, and finding intervention points.

• model-eisenhower

  Urgency/importance prioritizer using the four-quadrant matrix. Use when overwhelmed, everything feels urgent, or reactive mode is taking over.

• model-first-principles

  Assumption challenger who strips away conventions to find fundamental truths. Use when stuck in "how it's always been done" or need breakthrough thinking.

• model-inversion

  Failure mode analyst who asks "What would guarantee failure?" Use for risk prevention, planning important initiatives, and pre-mortems.

• model-occams-razor

  Simplest explanation finder who prefers theories with fewest assumptions. Use when multiple explanations seem plausible or debugging with many possible causes.

• model-one-thing

  Leverage identifier who finds the single action that makes everything else easier. Use when overwhelmed or seeking force multipliers.

• model-opportunity-cost

  Tradeoff analyst who reveals true costs by examining what you give up. Use for resource allocation decisions and when something feels "affordable.

• model-pareto

  Impact prioritizer who finds the vital 20% driving 80% of results. Use when too many things compete for attention or seeking maximum results with minimum effort.

• model-second-order

  Consequence chain analyst who thinks through effects of effects. Use for decisions with long-term implications or complex system interventions.

• model-six-hats

  Parallel thinking facilitator using De Bono's Six Thinking Hats. Use for group decisions, exploring ideas from multiple angles, or when emotions and logic need separation.

• model-swot

  Strategic position analyst mapping internal factors (strengths/weaknesses) and external factors (opportunities/threats). Use for competitive assessment and strategic planning.

• model-toc

  Theory of Constraints analyst using Current Reality Tree, Evaporating Cloud, and Future Reality Tree. Use for systemic problems with multiple symptoms, hidden conflicts, or when obvious fixes keep failing.

• model-via-negativa

  Simplification specialist who improves by subtraction rather than addition. Use when systems feel bloated or adding more isn't helping.

• research

  Authoritative research using Firecrawl MCP with rigorous source evaluation. Use PROACTIVELY when investigation is needed before action or when 3+ related searches would be needed. MUST BE USED when user says "research", "investigate", "find out about", or needs verified facts from credible, primary

• think-classifier

  Problem classification specialist who identifies problem type and characteristics for model selection. Use at start of /think:consider.

• think-clr-validator

  Adversarial logic validator using Goldratt's Categories of Legitimate Reservation. Validates causal links for TOC analyses.

• think-orchestrator

  Model selection and coordination specialist who selects mental models and plans their parallel or sequential execution.

• think-synthesizer

  Result synthesis and voting specialist who merges parallel agent outputs into unified insight with consensus scoring. Optionally stores insights to MCP memory.

• think-validator

  Adversarial validation specialist who stress-tests synthesized conclusions for blind spots and edge cases.

Similar plugins