deep-research

Multi-agent deep research decomposes your topic into MECE angles, runs parallel Opus sub-agents for investigation and gap analysis, and synthesizes findings into a cited document. This ensures exhaustive topic coverage with attributed sources and contradiction reporting.

MECE decomposition breaks a research question into 3–6 non-overlapping angles to ensure comprehensive investigation. This structured approach prevents missing important perspectives during multi-agent research and gap analysis.

To save comprehensive research reports into a vault document, instruct your AI to run the deep-research process on your topic. It generates a structured report with YAML frontmatter discovery metadata and outputs a final vault path for storage.

Deep research handles current web information by injecting today’s date into sub-agent prompts, ensuring near-2026 data retrieval. It also performs gap analysis to report contradictions and evaluate evidence quality across sources.

Yes, you can use parallel Opus sub-agents for gap analysis on any research subject requiring exhaustive coverage. The orchestration pipeline enforces source-quality rules and synthesizes multi-angle findings into a single vault-ready research document.

Limitations of multi-agent research include potential latency from running parallel Opus sub-agents and the necessity of structuring outputs into a specific vault document format with YAML frontmatter, which requires a compatible research vault setup.