update-landscape

name	update-landscape
description	This skill should be used when the user asks to "add a company", "add a relationship", "update an edge", "remove a node", "add more companies", "update the data", "add startups", "add X companies", or any request to modify the tech landscape graph data (nodes or edges). Also triggers on "who's missing", "what relationships are missing", or "update company info".

Update Tech Landscape

Manage companies (nodes) and relationships (edges) in the Silicon Valley tech landscape visualization. Data lives in data.js, loaded by index.html.

Data Location

• Data file: data.js — contains the DATA object with nodes and edges arrays
• Visualization: index.html — reads DATA from data.js, renders D3 force graph
• Schema reference: references/schema.md — full field definitions, valid types, constraints

Workflow: Adding Companies

1. Read data.js to see existing nodes and avoid duplicates
2. Read references/schema.md for the node schema
3. For each new company, determine:
   • id: lowercase, unique, no spaces (e.g., "databricks")
   • cluster: one of AI, Cloud, Social, Hardware, Enterprise, Fintech, Transport, Consumer
   • All required fields: name, founded, hq, ceo, description
4. Insert new node objects into the DATA.nodes array in data.js
5. Identify relationships between the new company and existing companies
6. Add corresponding edges (see below)

Workflow: Adding Relationships

1. Read data.js to verify both source and target node IDs exist
2. Read references/schema.md for the edge schema and valid types
3. For each relationship, determine:
   • type: ally, rival, investor, acquirer, lawsuit, or frenemy
   • strength: 1-5 scale (see schema for calibration)
   • startYear / endYear: when the relationship started and ended (null if ongoing)
   • summary: 1-3 factual sentences explaining the relationship
   • events: 2-6 timeline entries with year and description
4. Insert new edge objects into the DATA.edges array in data.js

Workflow: Updating Existing Data

1. Read data.js and locate the node or edge to update
2. Apply changes directly — update CEO names, descriptions, add new events to edges, etc.
3. When updating a company's cluster, verify no edges assume the old cluster for filtering

Workflow: Removing Data

1. When removing a node, also remove ALL edges that reference it as source or target
2. When removing an edge, no other cleanup is needed

Validation Rules

Before finalizing any data change, verify:

• Every edge source and target references an existing node id
• No duplicate node id values
• All required fields are present on every node and edge
• strength values are integers 1-5
• startYear <= 2026 for edges that should be visible by default
• type is one of the 6 valid relationship types
• cluster is one of the 8 valid industry clusters
• Event years are within the startYear-endYear range (or startYear-2026 if endYear is null)
• Descriptions and summaries use factual, verifiable information

Research Guidelines

When adding companies or relationships based on user requests:

• Use factual, well-known public information only
• Dates should be accurate (founding years, lawsuit dates, acquisition dates)
• CEO names should be current
• Relationship summaries should be neutral and factual, not opinionated
• When uncertain about a fact, note it and ask the user rather than guessing
• Prefer well-documented, notable relationships over obscure ones

Quick Reference

Add a company

Edit data.js → DATA.nodes array → insert { id, name, cluster, founded, hq, ceo, description }

Add a relationship

Edit data.js → DATA.edges array → insert { source, target, type, strength, startYear, endYear, summary, events }

Valid clusters

AI, Cloud, Social, Hardware, Enterprise, Fintech, Transport, Consumer

Valid relationship types

ally, rival, investor, acquirer, lawsuit, frenemy

Additional Resources

Reference Files

• references/schema.md — Complete field definitions, valid values, constraints, and performance notes