Academic Register — Security skill for Claude Code
Claude Code skill: audit & fix English academic prose register — guards BOTH poles (too-informal ↔ AI-inflated), so a 'make it formal' pass never creates AI-slop.
How to install Academic Register
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open fbdeme/academic-register and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Academic Register does
Claude Code skill: audit & fix English academic prose register — guards BOTH poles (too-informal ↔ AI-inflated), so a 'make it formal' pass never creates AI-slop. Severity rubric (R1–R10 + Hyland stance), deterministic register_scan.py, cited research brief. English sibling of korean-humanizer; pairs with prose-readability/stat-tone.
Alternatives in Security
- ThinkWatch — Self-hosted AI API and MCP gateway for organizations: SSO and RBAC, per-user identity for MCP tool calls, PII 817 ★
- Memory Self Review — Mine recent agent history (claude-mem + usage stats) for recurring failures and repeated patterns, audit MEMOR 297 ★
- SEO Audit Skill — English · 中文 可复用的单页面 SEO 审计 Agent Skill 235 ★
README
academic-register
A Claude Code skill. Fixes English academic/scientific prose that reads wrong to reviewers — either **(A) too informal** (contractions, idioms, phrasal verbs, "a lot of", rhetorical questions, punchy fragments) or **(B) AI-written/inflated** (delve, underscore, tricolons, "not only… but also", "it is worth noting", hedge clusters, promotional verbs). It measures with a deterministic scan, classifies violations by severity (S1/S2/S3) against a canonical rubric (R1–R10 + Hyland stance), and rewrites by **substituting** (not deleting) while preserving content, numbers, citations, and voice.
Its core is the **two-pole guard**: a naive "make it formal" pass pushes writing from A straight into B, so the both-pole AI-slop intersection is off-limits in *both* directions.
- `SKILL.md` — the skill (iron laws, severity, workflow, grades).
- `references/register-catalog.md` — full R1–R10 + stance + Do-NOT.
- `references/research_brief.md` — cited research foundation.
- `scripts/register_scan.py` — deterministic trigger/density scan (
--demo).
The English scholarly-register sibling of [`korean-humanizer`](https://github.com/fbdeme/claude-skills); pairs with [`prose-readability`](https://github.com/fbdeme/prose-readability) and [`stat-tone`](https://github.com/fbdeme/stat-tone). MIT.
Related Skills
Gaia Debt
Fix the tech-debt backlog, a single issue or a recommended related batch, highest severity then oldest first,
Evaluator Library
Dispatch a fresh-context evaluator against a domain rubric. Bundled domains: code-quality, ui-design, prose, s
Copy Crusade
Unleash parallel Copy Purist agents to audit copy at BOTH the text level (vague buttons, robotic errors) AND t
Prose Audit
Lint paper drafts for AI-tell prose patterns. Args: [--severity=warn] [--no-pencil] [--no-muriel] --no-native
Academic Paper Accuracy And Citation Audit
Use when: (1) performing a multi-pass LaTeX paper audit covering data consistency, cross-references, scientifi
Paper Style Guide
Claude Code skill: audit and rewrite academic ML/CS paper drafts. Pass A (style) + Pass B (submission checklis
Related Agents
Agy Stylist
Rewrites prose for style through Gemini via the agy CLI, in any register (legal, technical, academic, commerci
Sdd Context Keeper
Maintains informal project context (preferences, deferred decisions, stakeholder comments) that does not belon
Search Reducer
Read-only/advisory REDUCE role for the finite-computation solver layer in lean-proof-agents. Given a formal-co