deep-web-research

What problem does it solve?

This Skill eliminates one-off, low-confidence web searches and delivers rigorous, multi-source analysis suitable for high-stakes decisions by enforcing triangulation, source-quality scoring, contradiction checks, and explicit confidence labels.

Core Features & Use Cases

• Hypothesis-driven framing: convert a request into a primary question, 3–7 subquestions, and falsifiable hypotheses to guide search scope.
• Source-tiered evidence collection: prioritize Tier 1/2 sources, log URL, date, tier, and evidence strength, and require multiple strong citations for major claims.
• Contradiction testing and synthesis: surface disagreements, test counterevidence, and produce evidence-backed conclusions with confidence scores and recommendations.
• Parallel, structured workstreams: run up to four concurrent lanes for market size, customer signals, competition, and risks then centralize synthesis into a single decision-grade output.
• Use case example: perform market entry due diligence, vendor landscape comparisons, or technology risk assessments that directly inform go/no-go decisions.

Quick Start

Conduct deep research on the competitive landscape for entering the EV charging market over the next 12 months and deliver a full_report with sources, confidence scores, contradictions, and recommendations.