Paper-writing skill that tracks a draft's modification history across sessions via a per-draft journal file backed by git, and infers the draft's forward trajectory (not just past diffs). The flagship skill in this marketplace.
SEO improvement, recovery, and audit skill for existing sites — plus GEO (AI-search optimization for Perplexity, ChatGPT Search, Gemini), AEO (snippets/voice), and keyword analysis. Modes: IMPROVE (default — ground truth from the codebase and Search Console, prioritized fix backlog, shipped with regression tests), RECOVER (traffic drops, deindexing, migration damage), AUDIT (scored SEO/GEO/AEO report), plus direct implementation of named techniques from a 30-entry playbook (meta tags, schema markup, sitemaps, redirects, llms.txt, OG images, Core Web Vitals). Use when the user wants to improve SEO, fix rankings, diagnose a traffic drop, recover from a migration, audit a site, check AI search readiness, or target keywords.
Produce high-impact technical articles using the "Sewing and Reaping" methodology — authentic engineering exploration first, then narrative — with built-in SEO, GEO (AI search), and AEO (snippets/voice) scaffolding. Generates SVG diagrams inline or a JSON image spec for an image generator. Use when the user wants to write a technical blog post, engineering essay, case study, deep-dive, tutorial, postmortem, or launch announcement; to draft, outline, structure, optimize, audit, or review an article for a technical audience; to find the narrative kernel or angle in work they've built; to add diagrams or hero images to technical content; or to rewrite AI-sounding text so it reads like a real engineer wrote it (de-AI a draft, remove AI tells, humanize prose). Writes in a practical walkthrough voice with a hard ban list of AI tells. Trigger on "help me write about X", "draft a blog post", "write a tutorial", "find the story in this work", "make this rank on AI search", "this sounds like AI", or "review my draft".
Review a document as its readers would before they do: simulated readers who know only what a real reader knows walk or read it towards their own purpose, editors apply rubrics, a verifier per reviewer locates every quote and tries to refute the major claims, and a synthesis ranks what survives as predictions to test with real readers.
Rewrite drafts, articles, emails, Slack messages, proposals, or LLM-generated copy into Joel Azemar's voice so it reads human, not AI-generated: picks the register (public company copy or Joel's own messages), enforces the no-dash and no-invented-facts rules, cuts hedges and empty intensifiers, and returns paste-ready text plus a short note of what changed.
A Stop hook that checks the final message of each turn reads on its own for an owner who reads nothing else: no restating recent final messages, no hard-to-read phrasing, and no names or numbers left unexplained. A failing message goes back for a rewrite, at most twice per turn. Fails open with a visible warning. Claude Code only. No MCP server, no skills.
BinPar brand voice (es-ES): clinical sobriety in a conversational voice, naming, red lines, format guidance (web, blog, video, LinkedIn, documents, email, UI) and a review mode for any BinPar copy.
Writing for client deliverables: a copyedit pass for English and German reports, assessments, proposals, statements of work, emails and slide text that removes AI tells, fixes locale formats and consistency, keeps every number and commitment intact and reports what changed. Ships a deterministic prose lint with a compare mode.