Turns plain text or rough markdown into a polished, scannable article with frontmatter, titles and structure — without changing a single word of the original.
A three-mode (quick/normal/refined) translation pipeline for Claude Code with glossary merging, chunking, and parallel subagent drafting.
Adds an inner-dialogue step so the AI critiques its own draft and answers like a real person instead of a helpful assistant.
Drafts competitive funding proposals tailored to NSF, NIH, DOE, DARPA, and Taiwan NSTC review criteria and formatting rules.
Draft, revise, and audit scientific manuscripts with enforced evidence provenance, reporting-guideline coverage, and authorship/disclosure checks.
Diagnoses and strips AI writing tells from web-novel prose with minimal, function-preserving edits.
A staged pipeline that indexes a long novel's chapters, dissects the opening three chapters, summarizes every chapter, and synthesizes plot, emotion, pacing, characters and style reports.
Detects AI-writing tells in fiction drafts and rewrites them naturally with the smallest possible edits.
Converts Markdown into styled, inline-CSS HTML that is ready to paste into WeChat and other publishing platforms.
A P0–P7 research pipeline with parallel subagents, a locked citation registry, source-type governance and mandatory counter-review for traceable reports.
Verifies factual claims in your documents against authoritative sources and applies corrections only after your approval.
An execution protocol for the md2wechat CLI that turns Markdown into WeChat Official Account HTML, drafts, and image posts.
A read-only pipeline that turns a long novel or serialized web fiction into a traceable analysis layer: chapter index, adaptation-value triage, per-chapter function extraction, and episode candidates.
Turns novels, loglines, or multi-episode drafts into traceable short-drama development docs: creative brief, story engine, and episode map.
Rewrites AI-sounding text into human-sounding prose using countable rules for sentence rhythm, vocabulary, and punctuation.
Runs a forensic, evidence-cited scan of any text across nine AI-tell categories and returns a 0–27 score with a verdict and fixes.
Scrapes and analyses Chinese web-novel platform rankings (Qidian, Fanqie, Jinjiang, Qimao) to surface genre trends and actionable story-topic picks.
Reverse-engineers a novel you have already written (partial or finished) into a standard, resumable writing project with settings, outlines, and continuity tracking.
A runtime protocol that drives the md2wechat CLI to turn Markdown into WeChat Official Account HTML, with previews, covers, image posts, and gated draft creation.
An editorial playbook for deciding what humans must own, where AI may help, and how to keep AI-assisted content on-voice and honest.