Refine structure, wording and tone so existing copy is easier to understand.
Editing isn't rewriting. The job splits into seven focused passes, one dimension each: can the reader understand it; is the voice consistent; what does each claim mean for the buyer and what proves it; abstractions replaced with concrete scenes and numbers; the right emotion preserved; liability-hedging clichés cut. Each pass re-checks that earlier improvements survived.
It works on your actual copy — read from the workspace or Notes — and delivers an edited new file with a reason per change, the original left intact. Readability and similar measures are compared before and after so improvements are real, not cosmetic; when a claim needs checking it uses the app's own search, and unverifiable bits stay flagged instead of being smoothed.
Examples describe the request and the shape of the expected deliverable; they are not completed run results.
Don't rewrite this launch email — run the seven passes. Give me each pass's issues separately: clarity, voice, reader payoff, evidence…
Findings grouped per editing pass with rewrite suggestions and before/after readability measures, original untouched.
Every 'industry-leading' or 'massively improves' claim in this copy: mark ones with verifiable evidence, and rewrite the rest into honest, downgraded versions.
An edited copy with evidence tags and honest downgrades per claim, re-checked afterward to confirm no regression.
iOS · Android
iOS
Android
MIT
MIT (root LICENSE)
The pinned skill subtree and shipped references were audited; full applicable license and per-file adaptation/asset notes are preserved under references/licenses/ and references/PROVENANCE.md.
Publication commit 25e07dbeeb, audited source 84c524b6fb; catalog digest iOS 46196ba57ed7… · Android 16de89d019e2….
Version records ship with the signed catalog; the target and released versions are shown below.
