deep-context

# Deep Context Comprehensive cross-channel research on any topic. Searches every connected tool in parallel, then synthesizes findings into a structured briefing. Use this when you need the full picture on a project, feature, incident, or initiative. ## When to Activate - User says "deep context on X", "tell me everything about X" - User says "research X", "full context on X", "what do we know about X" - User needs to get up to speed on an unfamiliar topic quickly ## Input Required: - **Topic**: The subject to research (project name, feature, incident, concept, person) Optional: - **Time range**: How far back to search (default: 90 days) - **Depth**: "quick" (summary only) or "deep" (full timeline + all sources, default) ## Execution ### Context Check Before searching, read all four context files (`context/company.md`, `context/product.md`, `context/competitors.md`, `context/personas.md`) if they exist. Existing context informs search terms and avoids redundant research. After synthesis, offer to update context files with new knowledge discovered. ### Step 1: Expand Search Terms Before searching, generate 3-5 related search terms. Topics often have aliases, abbreviations, or related concepts. Example: "authentication redesign" expands to: "auth redesign", "login", "SSO", "identity", "auth v2" ### Step 2: Parallel Data Gathering (Fan-Out) Search ALL connected MCP sources in parallel using each search term. | Source | Method | Wha...