Sharpen a vague business question into a decision-oriented problem statement with clear scope

Run a full strategy consulting engagement — from problem definition through research, validation, synthesis, and executive-grade client report

Plan expert interviews and create structured interview guides mapped to research gaps

Deploy three research agents to investigate a topic — two independent analysts plus a validator

Independent research analyst (Thread A). Use this agent when the research skill dispatches parallel research workstreams to investigate a consulting hypothesis or client question. <example> Context: The research skill is investigating market dynamics for a client engagement user: "Research whether the client should enter the European EV charging market" assistant: "I'll dispatch analyst-alpha to research this from the market-and-demand angle." <commentary> Analyst-alpha takes the first independent research thread with a dynamically assigned angle tailored to this specific question. It works independently from analyst-bravo to reduce confirmation bias. </commentary> </example> <example> Context: The research skill needs evidence to test multiple hypothesis branches user: "Find evidence on whether private label expansion drives grocery margin recovery" assistant: "Deploying analyst-alpha to research private label economics and margin benchmarks." <commentary> Analyst-alpha receives a dynamically generated angle from the research orchestrator, ensuring the angle is relevant to the specific question rather than a generic category. </commentary> </example>

Independent research analyst (Thread B). Use this agent when the research skill dispatches parallel research workstreams to investigate a consulting hypothesis or client question. <example> Context: The research skill is investigating market dynamics for a client engagement user: "Research whether the client should enter the European EV charging market" assistant: "I'll dispatch analyst-bravo to research this from the competitive-landscape-and-regulatory angle." <commentary> Analyst-bravo takes the second independent research thread with a dynamically assigned angle tailored to this specific question. This deliberate separation ensures broader evidence coverage and reduces groupthink. </commentary> </example> <example> Context: The research skill needs evidence to test multiple hypothesis branches user: "Find evidence on whether private label expansion drives grocery margin recovery" assistant: "Deploying analyst-bravo to research retailer negotiation dynamics and branded manufacturer responses." <commentary> Analyst-bravo receives a dynamically generated angle from the research orchestrator, ensuring it covers different territory than analyst-alpha for this specific question. </commentary> </example>

Deep-dive research analyst that targets specific under-explored sub-dimensions identified after the initial two-analyst pass. Use this agent when the research skill dispatches a third research agent to fill gaps in the validated findings. <example> Context: The validator identified under-explored sub-dimensions after the initial research pass user: "Dispatch a deep dive on pricing granularity by segment and regional regulatory differences" assistant: "I'll dispatch analyst-deep to investigate these specific gaps in depth." <commentary> Analyst-deep receives the validated findings from the first two analysts and a targeted list of sub-dimensions to investigate. It goes deeper rather than broader. </commentary> </example>

Research validation agent that cross-checks findings from two independent research analysts for consistency, accuracy, and source quality. Use this agent after analyst-alpha and analyst-bravo have completed their research. <example> Context: Both research analysts have completed their investigation user: "Validate the research findings from both analysts" assistant: "I'll deploy the research-validator to cross-check both memos for consistency and accuracy." <commentary> The validator acts as a quality gate — it catches contradictions, flags weak sources, and produces a unified, trustworthy evidence base before synthesis begins. </commentary> </example> <example> Context: Research is complete and needs to be consolidated before the sense-check phase user: "Are these research findings reliable enough to build a storyline on?" assistant: "Let me run the research-validator to assess the evidence quality and flag any issues." <commentary> The validator determines whether the evidence base is strong enough to proceed or whether additional research is needed. </commentary> </example>

Conduct thorough, multi-tier research on a client question by dispatching two independent research agents and a validation agent. Use when someone asks to "research this topic", "find evidence for", "investigate this market", "gather data on", "what does the evidence say about", or when the analytical workflow reaches the research phase after problem definition. Also trigger when someone uploads client data or expert interview notes that need to be analyzed alongside public research.

Pressure-test research findings and analytical conclusions for robustness. Use when someone asks to "sense-check this", "pressure-test these findings", "challenge this analysis", "stress-test the argument", "play devil's advocate", "poke holes in this", or when the analytical workflow needs a quality gate between research and synthesis. Also trigger when someone presents a conclusion and wants it challenged before taking it to a client.

Synthesize research findings into a coherent storyline that answers the client question. Use when someone asks to "synthesize these findings", "build the storyline", "what's the narrative", "put this together into a story", "structure the argument", "what should we tell the client", or when the analytical workflow moves from sense-checked research to constructing the client-facing argument. Also trigger when someone has a collection of findings and needs help turning them into a logical, persuasive narrative.

Package synthesized findings into an executive-grade Word document (.docx) deliverable. Use when someone asks to "write the report", "create the deliverable", "produce the client document", "draft the final report", "package this into a document", or when the analytical workflow reaches the final delivery phase after synthesis. Also trigger when someone has a completed storyline and needs it turned into a professional, client-ready written report.

Run a full strategy consulting analytical engagement from problem definition through to a client-ready report. Use when someone asks to "run a full analysis", "do a consulting engagement on", "analyze this end to end", "help me with a client engagement", "do the full workflow", or presents a business problem that needs the complete Define → Research → Sense-Check → Synthesize → Deliver treatment. This is the orchestrator skill — it coordinates all other skills in the plugin to produce a comprehensive deliverable.

Plan, guide, and process expert interviews to confirm and enhance research findings. Use when someone asks to "prepare for expert interviews", "create an interview guide", "who should we interview", "process interview notes", "extract insights from interviews", or when the analytical workflow includes expert interviews after the research phase. Also trigger when someone uploads expert interview transcripts or notes that need to be analyzed and integrated into the research base.

Build a structured hypothesis tree with testable branches to guide analysis. Use when someone asks to "build a hypothesis tree", "structure this problem into hypotheses", "create an issue tree", "break this down MECE", or when a defined problem would benefit from structured decomposition before research begins. This skill is optional — not every problem requires a formal hypothesis tree. Skip it when the question is narrow enough that direct research is more efficient.

Define and scope a client problem into a sharp, decision-oriented question. Use when someone says "help me define this problem", "scope this engagement", "what's the real question here", "sharpen this problem statement", or provides a vague business challenge that needs structuring before analysis can begin. Also trigger when someone shares a client brief, RFP, or project description and needs help extracting the core analytical question.