Blind Taste Test announcement
Start with one headline, compare two options, then ship the stronger one. ContentForge gives you a grade, specific feedback, and rewrite help before you publish. Same input, same score, every time.
Score badges on Twitter/X, LinkedIn, Instagram, Threads, and Facebook while you write. Install free, works immediately.
Start with a weak headline, rewrite it, then check whether the new version actually earns the pass.
When the scorer gets a ranking right, the evidence lands here. When it misses, the miss is logged too.
Before the public calibration log fills up, the launch itself is already teaching us what people trust and what they ignore.
Blind Taste Test announcement
Broad API feature pitch
Technical explainer without fast proof
The landing page should stay focused on the scorer, but the technical path is still here if you want to wire it into your own workflow.
Use the compare and scoring endpoints when you want to rank drafts in code instead of by hand. The API is secondary on this page, but it remains the path for automation and paid usage.
The deterministic scoring logic is open source, readable, and runnable locally. If the core promise is “same input, same score,” the source needs to stay auditable.
Interpretable content intelligence — every deduction has a traceable rule, every score has an audit trail.
A ruler doesn't need a dataset to measure 12 inches — it just needs to be calibrated. ContentForge's heuristic engine gives the same score to the same input, every time. No variance. No hallucinations. Every deduction is itemised and traceable to a specific rule.
All 12 platform scorers are pure Python heuristics. No network calls, no model inference, no API quota consumed. AI (Ollama or Gemini) is reserved for generation endpoints — rewrites, hooks, subject lines.
The scoring logic is on GitHub. You can trace exactly why a post scored 74 and not 83 — every signal is a readable Python condition. When a client asks "why did this fail the quality gate?", you have a defensible answer, not "the AI said so."
Instant heuristic scoring in the browser. Score one draft or compare two before you publish.
Headline · 30–80 chars is optimal · add a number or question
Paste two drafts and see which one wins across platforms. The current defaults mirror an early launch lesson: concrete proof beats broad product description.
Ask GPT to score the same tweet twice. You'll get two different answers. That's the problem.
Try the live compare workflow, inspect the rules, then decide if you need the API or extension when approval lands.
Try The Live Scorer →Questions? captainarmoreddude@gmail.com
The live scorer on this page is free to try. The API is available separately through RapidAPI if you want programmatic access and higher-volume usage.
No. Text is scored in memory and discarded immediately. We do not log, store, or analyze your content after the response is sent. See our Terms of Use.
ContentForge runs on Render's free tier, which puts the server to sleep after inactivity. The first request wakes it up (~15-30 seconds). Subsequent requests are fast (<500ms). Paid infrastructure is on the roadmap.
The extension is live on the Chrome Web Store. It shows a floating score badge on Twitter/X, LinkedIn, Instagram, Threads, and Facebook while you write. You can also use the popup on any page to score content for all supported platforms.
The instant scoring endpoints use deterministic rule-based analysis (no LLM, no latency, no hallucinations). The AI endpoints use Ollama-backed language models for content generation, rewriting, and deep analysis.
Yes. The API still exists for compare mode, scoring, and automation. It is just no longer the main thing the landing page should sell while the product is still proving the core scorer.
Yes. ContentForge is open source: github.com/CaptainFredric/ContentForge. Every scoring rule is public and auditable.
The clearest pattern was that public calibration and before or after comparison got more trust than a giant feature list. The live notes are tracked in docs/reddit-launch-notes.md, and they are already shaping the product narrative.