Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/epicsagas/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/rules/epicsagas/velith/velith-pipeline)<a href="https://agentmods.dev/rules/epicsagas/velith/velith-pipeline"><img src="https://agentmods.dev/badge/rules/epicsagas/velith/velith-pipeline/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/epicsagas/velith/velith-pipeline"><img src="https://agentmods.dev/badge/rules/epicsagas/velith/velith-pipeline.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01381 | $0.01381 |
| Opus 5.5 | $0.00552 | $0.00552 |
| Sonnet 5.5 | $0.00276 | $0.00276 |
| Haiku 4.5 | $0.00138 | $0.00138 |
Grade A, and why
velith-pipeline 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 1mo ago.
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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Velith — Books to the Human-Quality Bar
Overview
Velith produces books a cold reader cannot distinguish from a competent human author's work. Six phases, twelve agents, one standard: skills/loom/quality-bar.md (five-axis rubric, the 2026 AI-tell taxonomy, the cold-read protocol). Read it before any writing or editing work.
Phase 0: Onboarding → Phase 1: Ideation → Phase 2: Outlining → Phase 3: Drafting → Phase 4: Editing → Phase 5: Publishing
Operating principles
- Full-manuscript context. Writing and editing agents read the entire manuscript so far, not summaries.
bible.mdis an index and ledger, not a substitute. - Sequential drafting for narrative work. Fiction, screenplay, game, memoir, narrative nonfiction: chapter N only after N-1 exists and its bible ledger is updated. Parallel (max 3) only for technical/reference/academic chapters the outline DAG proves independent.
- Voice lock before volume. No drafting beyond the sample chapter until the author approves the voice; the lock is recorded in
STYLE.md. - Draft, critique, revise. Every chapter is cold-critiqued with quoted lines and revised before it is saved. Editing stages rewrite the manuscript in place.
- Grounded claims. Nonfiction, technical, academic: fact-check before line edit. Unverifiable numbers, quotes, studies, citations are removed, not softened.
- Reader gate. Phase 4 ends with a cold read by simulated target readers (
beta-reader). Publishing requiresverdict: PASSinedits/readiness-report.mdor an explicit author override. - One look per book.
art-bible.md; prompts compiled byvelith.mjs images compile; diagrams and drawings code-rendered; vision QA on every image; look lock before volume. - Author checkpoints. Concept selection, outline approval, voice lock, look lock, developmental restructures, readiness verdict. Everything else runs without asking.
Phase router
| State | Route |
|---|---|
No PRD.md |
Phase 0 /book-init |
PRD.md, no ideation.md/outline.md |
Phase 1 /book-ideation |
ideation.md, outline missing or without ## Approval |
Phase 2 /book-outline |
Approved outline + bible.md, no drafts, or STYLE.md ## Voice Lock pending |
Phase 3 voice lock, then drafting |
| Drafts incomplete | Phase 3 /book-draft (resume) |
| All drafts, no readiness PASS | Phase 4 /book-edit |
Readiness PASS, no publish/book.epub |
Phase 5 /book-publish |
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.
- 1mo ago Changed · -39 lines · +217 tokens per session 0763616a1713
- 2mo ago First seen · 125 lines · 1,164 tokens per session scan A 499adad31e2d
velith-pipeline is a cursor rule published in the GitHub repository epicsagas/Velith (40 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,381 tokens to every session, about $0.0055 per session on Opus 5.5. 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 cursor rules, from other repositories
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
prefer-assertions-over-defensive-checks
Prefer assertions over defensive checks when data is guaranteed to be valid.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
config-validation-and-normalization
Config validation and normalization responsibilities (defineConfig normalizes, the orm section schema validates and resolves paths, the loader reports per-section diagnostics).