Init research vault: structure, template, Bases config

- Per-topic directories with index.md entry point (mirrors idea-dump pattern)
- Frontmatter: title, status, category, tags, created, updated, origin
- Status lifecycle: exploring -> active -> concluded | shelved
- Categories: market, technical, competitive, legal, scientific, exploratory
- Obsidian Bases table view (bases/research-base.md)
- Template for new research topics (templates/research-template.md)
- Quartz-ready for future publication as research.rasyidanaf.com

README.md

@@ -0,0 +1,85 @@
+# Research Vault
+
+A vault for structured research — investigation, analysis, and synthesis across topics.
+
+## Workflow
+
+1. **Discuss** research questions in Buzz channels — `#research`, `#general`, wherever the question arises.
+2. **Declare** — when a research thread is worth keeping, @hermes-agent logs it here as a new topic directory.
+3. **Investigate** — the agent populates `index.md` with findings and creates detail files as the research grows.
+4. **Synthesize** — when research concludes, the `index.md` summary is finalized and status moves to `concluded`.
+
+## Structure
+
+Each research topic is a **directory** at the vault root with an `index.md` entry point:
+
+```
+<topic-slug>/
+  index.md              ← frontmatter + summary + key findings (the entry point)
+  sources.md            ← links, references, data sources
+  analysis.md           ← deep analysis, findings
+  methodology.md        ← how the research was conducted
+  notes.md              ← working notes, raw observations
+  summary.md            ← final synthesis (when concluded)
+  next-steps.md         ← follow-up questions, gaps
+```
+
+- **Small topics**: just `index.md` alone in the directory.
+- **Big topics**: split into detail files as they grow. Link them from `index.md` with `[[wikilinks]]`.
+- Detail file names are up to you — `sources`, `analysis`, `methodology`, `notes`, `summary`, `next-steps` are common patterns.
+
+### Frontmatter (in `index.md`)
+
+```yaml
+---
+title: One-line summary
+status: exploring | active | concluded | shelved
+category: market | technical | competitive | legal | scientific | exploratory
+tags: []
+created: YYYY-MM-DD
+updated: YYYY-MM-DD
+origin: buzz://<channel-uuid>/<event-id>
+---
+```
+
+### Categories
+
+Categories are in frontmatter, not folders — every topic lives at the vault root.
+
+| Category | Kind of research |
+|----------|-----------------|
+| `market` | Market analysis, sizing, trends |
+| `technical` | Technology evaluation, architecture research, tool comparison |
+| `competitive` | Competitor analysis, landscape mapping |
+| `legal` | Regulatory, compliance, legal landscape |
+| `scientific` | Literature review, paper analysis |
+| `exploratory` | Open-ended investigation, "what is X" |
+
+### Status Lifecycle
+
+```
+exploring → active → concluded
+                 ↘ shelved
+```
+
+- `exploring` — just started, gathering sources, scoping the question
+- `active` — ongoing research, regularly updated
+- `concluded` — findings synthesized, research complete
+- `shelved` — parked, may revisit
+
+## Bases
+
+Open `Research Base` (in `bases/research-base.md`) to see all topics as a filterable, sortable table — by status, category, tags, or date. Only `index.md` files appear; detail files are excluded via the `file.name.is("index")` filter.
+
+## Graph View
+
+Wikilinks between detail files within a topic, and between topics via shared tags, create a knowledge graph visible in Obsidian's graph view. Cross-links to [[index|Idea Dump]] entries (e.g., research that feeds a product idea) work natively since both vaults live under the same Obsidian vault.
+
+## Publishing (Future)
+
+The vault is structured for eventual publication as a public knowledge base:
+
+- **Quartz** (recommended) — free static site generator for Obsidian vaults. Preserves wikilinks, backlinks, graph view, and full-text search. Self-hostable via NixOS.
+- **Obsidian Publish** — paid, simplest, but not self-hosted.
+
+The `index.md` + wikilink structure means the published site will render as a connected knowledge graph, not a flat list of documents.