About & method.

Lantern is what I'd rather have read instead. One page, already filtered. An agent that read the week of research for me, dropped the duplicates and the disguised pitches, scored the rest on what matters — conversion relevance, evidence quality, actionability — and handed me a short selection with honest summaries. No marketing, no rephrased teaser.

I build it solo, evenings and weekends. It's an author project, not a startup. That means three things:

• No ads, no trackers. Lantern doesn't follow you, doesn't sell your data, and loads zero third-party script.
• No growth hacking. Ironic for a CRO site, but: no push notif, no gamification, no "streak". You read when you want.
• Explicit curation. Sources are public, scoring is documented below, and the AI never reorders your reading: the feed is chronological by default, score sort is opt-in.

Every article that lands in your feed traversed five steps. Most take a few seconds; LLM scoring is the bottleneck. The pipeline runs once a day — a chosen rhythm, not a technical limit: the ~10 active sources don't publish around the clock, one daily pass is plenty.

+-- 01 -----+    +-- 02 ------+    +-- 03 ------+    +-- 04 -----+    +-- 05 -----+
|  FETCH    | -> |  EXTRACT   | -> |  DEDUPE    | -> |  SCORE    | -> |  PUBLISH  |
+-----------+    +------------+    +------------+    +-----------+    +-----------+
rss-parser       readability       simhash 64b       gpt-5-nano     ranking + feed

What follows details each step — what it gets, what it produces, what can fail.

Scoring is the only place where AI decides something on your behalf. To keep it honest, here's the exact system prompt and the grid it applies. And you can always ignore the score: the feed is chronological by default — score sort and the min-score filter are opt-in.

The grid

• 5 ★ — Conversion core. Hard numbers, directly transferable to a real page or funnel. Study, before/after experiment, reproducible teardown.
• 4 ★ — Very good. Real evidence (a quantified case, explicit method) and applicable.
• 3 ★ — Solid but adjacent or familiar. Argued, one good example.
• 2 ★ — Marginal or promotional. Substanceless teaser, tool pitch without data.
• 1 ★ — Noise. Off-topic, claim without evidence, AI-generated prose.

Publish bar: the final score is the weighted composite of the three axes (0.4·relevance + 0.3·rigor + 0.3·actionability, 1.0–5.0 scale). An article passes when that composite is ≥ 3.5 / 5. Below that, it stays in the database (audit, calibration) but doesn't reach the public feed.

The system prompt (excerpt)

The prompt runs in French (the scoring model reads both, the rubric is authored in French). Translated here for reference:

# role: CRO / growth research evaluator
You are a research evaluator for a CRO / growth practitioner who
builds and grows a product (conversion, pricing, experimentation,
UX, tooling). You output a structured score, not a marketing recap.

# 3 axes, integers 1-5
- RELEVANCE       Useful to someone growing a product?
                1 = off-topic · 5 = conversion/pricing core,
                directly transferable to a page/funnel.
                AI GATE: AI content is relevant ONLY if it ties
                AI to a measured conversion/growth lever (test,
                funnel, pricing, quantified uplift). Otherwise
                (model review, "how I AI" tutorial, vendor agent
                feature announce) → 1-2.
- RIGOR           Thesis backed by data, method, a real test — or
                promo? A purely promotional teaser OR a post where
                a vendor pitches its own product with unverifiable
                internal numbers caps at 2, even from a reputable
                source.
- ACTIONABILITY   Applicable to a real page / a test this week?
                1 = theoretical · 5 = ready-to-use method.

# adjustments (the LLM applies them itself)
- anti-patterns (-1): empty listicle, disguised vendor content,
uplift claimed without numbers, AI boilerplate prose.
- bonus (+1): quantified before->after, quant research with method,
real hands-on report from a test.

Strict JSON output: { depth, novelty, applicability, type,
tags, summary, reasons }

The three axes are weighted in code (0.4 relevance · 0.3 rigor · 0.3 actionability) into the final score — they're stored under their legacy JSON keys depth / novelty / applicability, which is why the prompt output above uses those names. The prompt is versioned and frozen to stay comparable week to week; when we bump it (rare), it's noted in the changelog.

What I don't do

• No per-user personalization. Everyone sees the same editorial score. Lantern doesn't learn from your clicks.
• No A/B testing on titles. The displayed title is the source's, period.
• No commercial boost. No source pays to appear. No source is excluded for criticizing us.