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.
npx skills add wellwelwel/lagune --skill writergit clone --depth 1 https://github.com/wellwelwel/laguneWrote 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/skills/wellwelwel/lagune/writer)<a href="https://agentmods.dev/skills/wellwelwel/lagune/writer"><img src="https://agentmods.dev/badge/skills/wellwelwel/lagune/writer/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/skills/wellwelwel/lagune/writer"><img src="https://agentmods.dev/badge/skills/wellwelwel/lagune/writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00035 | $0.00969 |
| Opus 5 | $0.00017 | $0.00485 |
| Sonnet 5 | $0.00007 | $0.00194 |
| Haiku 4.5 | $0.00003 | $0.00097 |
Grade A, and why
writer 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 10d 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lagune writing
This skill is the specialized, authoritative description of how Lagune's prose is written: the language and style conventions, the plain-language rule, and the reconcile discipline that keeps artifacts living. Consult it before writing or editing any artifact, the charter, command text, a doc, a comment, or a user-facing message.
Language and style
- Avoid em dashes or semicolons where they are not grammatically necessary. Where one would go, for example, use the mark the sentence actually calls for, usually a comma, colon, or parentheses.
- Write in complete sentences, each with a beginning, middle, and end, instead of chopping one thought into fragments. A semicolon, an em dash, or a period used to cut a sentence short is the same habit wearing different punctuation, so the fix is to compose the sentence, not to swap the mark that split it.
- When the document already exists, read it and study its existing style before you edit, then write in that same voice. When it is entirely new, read a similar or equivalent file to learn the pattern first.
- Avoid writing a file path or pasting a code snippet into a document. A flat document does not track the source, so a moved file or a changed snippet leaves a stale, broken reference.
Rules
- All project content is written in English (docs, code, comments, command text, user messages), regardless of the language used in chat.
- Less is more: cut anything that adds words without meaning: throat-clearing openers that delay the point, reinforcement triads where one word does the job, framing sentences that announce content instead of being it, and padding that stretches a thought to look thorough. Stop once the point lands.
- Remove before you add: before editing existing text, first remove everything that is no longer valid, then adapt and update what changed, and add new text only when it is genuinely needed.
- Redundancy is forbidden: judged across the whole document, not one sentence at a time. A sentence clean on its own still counts as redundant when its meaning already lives elsewhere. Say each thing once, in the one place it belongs.
- Describe intent, not implementation: do not restate logic, cite specific variable or method names, or re-explain how something is built. A rename or a refactor silently turns that prose false, so say what it does and why, and let the code stay the single source of how.
- Never break lines by hand in the middle of the prose: line width, wrapping, and formatting are Prettier's job, not yours.
- Use straight quotes and apostrophes (
'and"), never their curly forms, and avoid any character a standard keyboard lacks.
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.
- 10d ago First seen · 53 lines · 35 tokens per session scan A d13f9705d2fa
writer is a skill published in the GitHub repository wellwelwel/lagune (148 stars, last pushed 10d ago), licensed MIT. It adds 35 tokens to every session and 969 once invoked, about $0.0002 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 skills, from other repositories
firebase-messaging
Use when setting up Firebase Cloud Messaging, managing permissions and tokens, handling background/foreground notification taps, or dispatching messages server-side (HTTP v1).
octocode-roast
Use when code needs a blunt evidence-backed roast or memorable critique: smell inventory, debt ranking, hot-path autopsy, savage/diff review, security or performance sins, or practical redemption paths. Phrases like roast this, brutal review, top sins, cleanup debt. Polite evidence-first PR review → octocode-research.
110-java-maven-best-practices
Use when you need to review, improve, or troubleshoot a Maven pom.xml file — including dependency management with BOMs, plugin configuration, version centralization, multi-module project structure, build profiles, or any situation where you want to align your Maven setup with industry best practices. This should…
sanity-best-practices
Sanity development best practices for schema design, GROQ queries, TypeGen, Visual Editing, images, Portable Text, Studio structure, localization, migrations, Sanity Functions, webhooks, Blueprints, and framework integrations such as Next.js, Nuxt, Astro, Remix, SvelteKit, Angular, Hydrogen, and the App SDK. Use this…
023-assumption-analysis
Use when a framed and root-caused problem needs its assumptions made explicit before design or planning begins — explicit Assumptions, Unknowns, and a Validation plan. This should trigger when an issue's Assumption Analysis point of view needs evaluation, or when a maintainer directly asks to surface hidden…
122-java-type-design
Use when you need to review, improve, or refactor Java code for type design quality — including establishing clear type hierarchies, applying consistent naming conventions, eliminating primitive obsession with domain-specific value objects, leveraging generic type parameters, creating type-safe wrappers, designing…