- meiliSearch now runs Meili hybrid (semantic+keyword, SEMANTIC_RATIO=0.5) with
a keyword-only fallback so search never hard-fails if the embedder hiccups
- Workers AI bge-small-en-v1.5 embedder configured on Meili (embeds on ingest)
- optional Claude reranker (?rerank=1) reorders/denoises the top 20 -> 10
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The crawl stalled when two crawler processes briefly overlapped on the same
WAL SQLite frontier (node:sqlite is synchronous, so a blocked write lock froze
the event loop). Fixes:
- run the crawler as a single-instance systemd service (search-crawler.service)
with CPUQuota=150% + MemoryMax=1200M so it can't starve a shared VM
- add timeouts to all Meilisearch calls + resilient flush (an untimed hang was
the amplifier that wedged every worker)
- crawl-diskguard.sh + cron: auto-stop the crawl and ntfy-alert if disk < 800MB
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
API: /answer retrieves top-6 from Meili and has Claude (haiku-4.5) synthesize
a 2-4 sentence cited answer over ONLY those sources (no hallucinated facts/URLs),
with a 1h in-memory cache. Frontend: async answer card above results with inline
[n] citations linking to sources — loads without blocking the results render.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace homepage-only crawling (the reason the index had ~9k docs and
useless results) with a real crawler:
- SQLite frontier (node:sqlite) — durable queue, seen-set, per-domain caps
- follows internal links to configurable depth, seeds from sitemap.xml
- robots.txt compliance + per-host politeness delay
- per-URL doc IDs (old id=domain overwrote every page of a site)
- link graph: cross-domain in-degree recorded as an authority signal
- authority score field + 'score:desc' ranking rule (relevance-first, authority tiebreak)
- body trimmed for the disk-bound VM
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A deployed Worker's fetch() cannot reach the API on :3000, so publish it at
https://search-api.theradicalparty.com through a Cloudflare Tunnel. Point the
frontend API_URL at HTTPS/443, add cloudflared install to install-vm.sh
(driven by CLOUDFLARED_TOKEN), thread the token through deploy.sh, and add
docs/DEPLOY.md with the full tunnel + DNS setup.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Crawler (Majestic Million → undici → cheerio → Meilisearch), Hono search
API with P2P peer federation, Cloudflare Worker frontend, VM provisioning
scripts. Adds README, .gitignore, and moves VM credentials out of deploy.sh
into a git-ignored deploy.env.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
