Orbit web apps
Orbit web apps
Paste any content. See which AI-drafted patterns are costing it points — and get a rewrite prompt you can drop into Claude.
The slop detector runs heuristic pattern-matching based on the anti-slop-editor skill that ships inside Orbit. It catches the specific phrases and structural tells that make content read as AI-generated, scores the draft out of 100, and gives you a rewrite prompt that activates Orbit's editing pass inside Claude.
The detection work is cheap, deterministic, and doesn't need an LLM. Scanning for 'it's important to note' or 'ultimately' doesn't require intelligence — it requires a dictionary. Keeping the detector local means no API costs, no round-trip latency, and no privacy footprint for whatever you paste in.
The rewrite is a different job. For that, the detector hands you a prompt pre-loaded with the findings and a reference to the anti-slop-editor skill. You drop it into Claude and get a version that applies the fixes — which Claude is genuinely better at than a rules engine would be.
Most AI-content tools are built to pass AI-detection services. This one is built to produce writing that doesn't read like AI slop — which is a different problem. The target is sharp, specific, voiced content. A draft can be entirely human-written and still score badly if it's padded with 'leverage the synergy'-style filler.
The categories — language, structure, substance, redundancy, audience — come from the editorial principles Orbit applies across every content-producing skill. If you use Orbit to generate blog drafts or FAQ content, you're already getting output filtered through the same pass this detector runs on paste-in content.
Score 85+: ship it. A few small tweaks at most. The detector probably flagged one or two minor items; your judgment wins.
Score 70–84: edit in place. Most of the findings are specific phrases you can swap or cut. Don't send the content through a full rewrite — you'll lose voice that the draft already has.
Score 50–69: consider a rewrite. The draft has enough patterns that editing each one is slower than asking Claude for a tighter version. Copy the rewrite prompt, paste into Claude, compare the output to your draft.
Under 50: rewrite from scratch. The draft is reading as mass-produced AI writing. No amount of phrase-swapping will save it — the structural and substance issues need a fresh pass.
Built into Orbit
Orbit ships with the anti-slop-editor skill as a built-in editorial pass. Every content-producing tool (copy framework, email production, program briefs) runs output through it automatically so generated content defaults to sharper and less generic.
Get Orbit freeA heuristic AI-content detector that scans blog drafts, FAQ pages, landing copy, and any prose for the patterns that make content read as AI-drafted — empty openers, filler transitions, non-opinion endings, corporate jargon, repetitive sentence openings, and fake completeness. Scores the draft out of 100 and returns a copy-paste rewrite prompt that activates the anti-slop-editor skill inside Claude.
Content marketers, lifecycle marketers, CRM leads, and anyone publishing written content regularly — especially teams using AI to draft and wanting a quality gate before things ship.
Built into Orbit for Claude
The anti-slop-editor skill runs as a silent final pass on every content-producing tool inside Orbit — copy framework, email production, program briefs, onboarding flows. Install Orbit in Claude Desktop and generated content defaults to sharper, less generic, more specific without asking.