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 epicsagas/Velith --skill book-poetrygit 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/skills/epicsagas/velith/book-poetry)<a href="https://agentmods.dev/skills/epicsagas/velith/book-poetry"><img src="https://agentmods.dev/badge/skills/epicsagas/velith/book-poetry/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/epicsagas/velith/book-poetry"><img src="https://agentmods.dev/badge/skills/epicsagas/velith/book-poetry.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 | $0.00046 | $0.01090 |
| Opus 5 | $0.00023 | $0.00545 |
| Sonnet 5 | $0.00009 | $0.00218 |
| Haiku 4.5 | $0.00005 | $0.00109 |
Grade A, and why
book-poetry 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 5d 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Poetry
A poem is the one form where the reader will forgive nothing. Every word is load-bearing. Machine poetry fails by being competent: correct images, reasonable line breaks, a closing turn, and nothing at stake.
Form
Choose per poem, not per collection, unless the collection is a formal sequence.
- Sonnet: 14 lines, volta; Petrarchan (8+6) or Shakespearean (4+4+4+2). Meter loosened in contemporary practice but the turn is not optional.
- Villanelle: 19 lines, two refrains. The refrains must change meaning through repetition or the form is empty.
- Ghazal: couplets, radif, the poet's name in the last couplet. Each couplet autonomous.
- Haiku / 하이쿠 / 俳句: cut, season, concrete image. 5-7-5 is a guideline in English; in Japanese it is the form.
- 시조: 3장 6구 45자 내외, 종장 첫 구 3음절 고정. Modern 시조 flexes everything but the 종장.
- Free verse: no fixed meter; the line break is the form. Every break must do something: emphasis, delay, double meaning, breath.
- Prose poem: no line breaks; density and sentence rhythm carry it.
The line
- A line is a unit of attention. What ends the line is emphasized. What starts it is emphasized. The middle is where you hide things.
- Enjambment creates speed and double readings; end-stop creates weight. A poem that only enjambs has no floor.
- Short lines intensify; long lines expand. Uniform line length is the first tell.
- Read every line aloud (in your head, slowly). Where the breath fights the line, decide whether that is the point.
Image and sound
- One central image per poem, pushed further than the reader expects.
- Concrete over abstract, always. "Grief" is not an image. The chair nobody moved is.
- Sound: consonant clusters, vowel length, stress patterns even in free verse. Assonance and consonance are more useful than end-rhyme in contemporary work.
- Metaphor that explains itself is deleted. The reader makes the link or does not.
- Korean: 음보 (3·4조, 4·4조, 7·5조) as a resource, not a rule; 시어 chosen for 소리 as well as 뜻; 한자어 vs 고유어 texture decided per poem.
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.
- 5d ago Changed · +43 lines · +14 tokens per session a04ac73cd9e3
- 10d ago First seen · 29 lines · 32 tokens per session scan A bf5d1698d397
book-poetry is a skill published in the GitHub repository epicsagas/Velith (35 stars, last pushed 5d ago), licensed Apache-2.0. It adds 46 tokens to every session and 1,090 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
ai-native-software-dev
Use when the user asks to build, change, modernize, or reconstruct non-trivial software with an AI-native workflow, including greenfield development, work in an existing codebase, or authorized reference-based product reconstruction; also use when lifecycle planning, architecture, implementation, testing, review…
akf-trust-metadata
The AI native file format. EXIF for AI — stamps every file with trust scores, source provenance, and compliance metadata. Embeds into 20+ formats (DOCX, PDF, images, code). EU AI Act, SOX, HIPAA auditing.
short-drama
A production workflow for making multi-episode AI short dramas, covering story planning, character consistency, scripts, generated video shots, voices, sound, subtitles, and final assembly.
paper-explainer
A research-content workflow that turns an arXiv paper or a local PDF into an explanation for a video or written article. arXiv is a public online archive of research papers.
multi-voice-dubbing
A script-to-audio workflow that gives different characters different voices and emotions, then combines their lines into one audio track with subtitles naming each speaker.
skill-competitor-analysis
A research process for studying how competing social-media accounts choose topics, format posts, publish, and engage with followers. It turns publicly available competitor activity into a content-strategy report.