alivemy-context-graph

# Map Visual overview of the entire ALIVE world — connections, clusters, health. Not a list (that's the tree in alive:world). Not a search (that's alive:search-world). Map is spatial — it shows how everything connects and where the energy is. --- ## What It Does ### 1. Generate the World Index Run `generate-index.py` from the plugin to walk all walnuts and collect frontmatter. Always use the plugin cache copy — never copy scripts to the world (world-local copies drift from the plugin and produce stale data). ```bash python3 "$ALIVE_PLUGIN_ROOT/scripts/generate-index.py" "$WORLD_ROOT" ``` The script walks all directories for `key.md` files, handling both `_kernel/key.md` (current structure) and root-level `key.md` (flat/legacy structure). It deduplicates entries and skips template walnuts. **Collected data per walnut:** - Name, type, goal, phase, rhythm, tags - People (from `key.md` `people:` field — multi-line list) - Links (from `key.md` `links:` field — wikilink extraction) - Parent (from `key.md` `parent:` field) - Bundles (detected by `*/context.manifest.yaml` in walnut root — name, status, goal) - Task counts (from `_kernel/now.json` if present) - Last updated (from `_kernel/now.json` `updated` field) **Projection tiers:** The graph supports three projection levels for controlling visual density: - **Tier 1 — Walnuts only** — just the walnut nodes and their connections. Fast, clean overview. - **Tier 2 — Walnuts + People** — adds people nodes that bridge multip...