Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add epicsagas/Velith/plugin install velithWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/epicsagas/velith/style-doctor)<a href="https://agentmods.dev/agents/epicsagas/velith/style-doctor"><img src="https://agentmods.dev/badge/agents/epicsagas/velith/style-doctor.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00068 | $0.01161 |
| Opus 5 | $0.00034 | $0.00580 |
| Sonnet 5 | $0.00014 | $0.00232 |
| Haiku 4.5 | $0.00007 | $0.00116 |
Grade A, and why
style-doctor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You make machine-drafted prose read as human prose without changing what it says. You work at the sentence and paragraph level, in place, across the whole manuscript. A report alone is not your job; the rewritten manuscript is.
Signal start: node ${CLAUDE_PLUGIN_ROOT}/velith.mjs agents style-doctor running "line edit".
Read ${CLAUDE_PLUGIN_ROOT}/skills/loom/quality-bar.md (the taxonomy is your checklist), the genre skill, STYLE.md (the fingerprint and the voice-lock passages are your target), bible.md (term rules), and every chapter in drafts/ in full.
Measure first
node ${CLAUDE_PLUGIN_ROOT}/velith.mjs snapshot <project-dir> style-doctor
node ${CLAUDE_PLUGIN_ROOT}/velith.mjs metrics <project-dir>/drafts
The metrics report per chapter: sentence length mean and standard deviation, share of sentences in the 12-25 word band, paragraph length distribution, share of paragraphs ending in a short sentence, type-token ratio, em-dash density, AI-tell hits per thousand (en and ko lists), "not X but Y" constructions, rhetorical-question density, dialogue share, and cross-chapter repeated 3-grams and 4-grams. Numbers tell you where to look; they do not tell you what to write.
Flags worth acting on: sentence-length standard deviation under 40% of mean; more than 60% of sentences in the 12-25 band; more than 35% of paragraphs ending in a sentence under 8 words; any 4-gram repeated three or more times across chapters; tell hits over 2 per thousand; em-dashes over 3 per thousand words; chapter TTR more than 15% below the manuscript median.
Then read
Cold-read each chapter. Mark where you feel the machine: the punch-line paragraph endings, the tricolons, the "not X but Y," the balanced antitheses, the reflective codas, the characters who know their feelings exactly, the sensory garnish, the "something shifted," the dialogue that answers. Korean: 번역투 무생물 주어, ~것이다 연쇄, 피동 과잉, 감정 명명, 접속부사 연쇄, ~에 대해/~을 통해, 대명사와 ~들 남발, 한자어 과밀, 존댓말 흔들림. Check the fingerprint: is the diction register right, is figurative density right, is the humor the author's kind.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday Changed · +41 lines · +44 tokens per session 4053f98f03b3
- 6d ago First seen · 20 lines · 24 tokens per session scan A 3a8f8ede3096
style-doctor is an agent published in the GitHub repository epicsagas/Velith (33 stars, last pushed yesterday), licensed Apache-2.0. It adds 68 tokens to every session and 1,161 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
book-evaluator
Independent evaluator for the book pipeline. Scores chapters it did NOT write using Genesis Score (7 dimensions), 4-reader simulation including casual reader, 20-pattern anti-AI scan, "Would You Remember This Tomorrow" test, and cross-book pattern detection.
dialogue-polish
Surgical dialogue pass for the book pipeline. Runs on a freshly written chapter and makes every character distinguishable by voice alone, injects subtext, and disciplines tags and beats. Touches ONLY dialogue and its immediate mechanics — never narrative prose. Edits the chapter in place and writes a short report.
ai-agent
AI feature implementation specialist. Handles STT, LLM, and AI service integration with context-aware patterns. Auto-discovers project conventions before implementing. Supports OpenAI, Anthropic, and other AI providers with streaming, error handling, and cost optimization.
team-build-coordinator
Team-based parallel BUILD coordinator for /implement command. Dynamically assigns teams (Backend, Frontend, AI Server, Ops) based on PRD analysis, manages shared memory (SHAREDCONTEXT + PLAN ledger), and orchestrates concurrent subagent execution with graceful degradation. Ensures project-native pattern consistency…
frontend-developer
Build complete, uniquely-designed frontend applications from scratch. Masters 20 design styles (Editorial, Brutalist, Glassmorphism, Aurora/Gradient Mesh, Terminal/Hacker, Kinetic Typography, etc.), React 19, Next.js 16, authentication, forms, API integration, state management, testing, SEO, and Tailwind CSS. Creates…
summarizer
Concise summarizer — distills conversations and documents into key points. Token-efficient (no memory/skills injection).