Borrowing it
Nothing to install: this file belongs to slowww-ai/second-brain. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/slowww-ai/second-brain/main/.claude/skills/shortwrite/SKILL.mdgit clone --depth 1 https://github.com/slowww-ai/second-brainWrote 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/slowww-ai/second-brain/shortwrite)<a href="https://agentmods.dev/skills/slowww-ai/second-brain/shortwrite"><img src="https://agentmods.dev/badge/skills/slowww-ai/second-brain/shortwrite/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/slowww-ai/second-brain/shortwrite"><img src="https://agentmods.dev/badge/skills/slowww-ai/second-brain/shortwrite.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 | $0.00157 | $0.00950 |
| Opus 5.5 | $0.00063 | $0.00380 |
| Sonnet 5.5 | $0.00031 | $0.00190 |
| Haiku 4.5 | $0.00016 | $0.00095 |
Grade A, and why
shortwrite 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shortwrite — long source → short angled piece
shortwrite.html (repo root) is a two-pane canvas:
- Left = SOURCE — a long article (e.g.
wiki/outputs/ko/skillopt-guide.md), opened read-only as reference, with a TOC. - Right = DRAFT — the new short piece, edited in the browser and saved to its own
.mdvia the File System Access API.
The user and Claude meet through the draft file; the source file is shared read-only context for both. Turn-based, like [[cowrite]].
The draft file contract
Every short draft carries this frontmatter — the angle and hook_type fields are the whole point:
---
title:
angle: # the specific perspective/hook this piece takes
hook_type: # 고백+깨달음 / 숫자+반전 / 변신 스토리 / Listicle / 정면반박
source: # filename of the long source article
format: short
status: draft
---
The loop
- Confirm both files. Identify the long source and the short draft for the session. Read the
source:field in the draft frontmatter; if set, that's the long article to compress from. - When the user says "이 관점으로 써줘 / 다듬어줘":
- Read the source article in full — it's the ground truth for facts.
- Read the draft — especially
angleandhook_type. The angle decides what to keep and what to cut. One angle per piece. - Write the short draft so it pulls only what serves that angle from the source. Don't restate the whole source; a short piece makes ONE point well.
- Apply the SNS hook playbook — the canonical framework at
.claude/skills/_shared/viral-hooks.md(read it for the 5 archetypes + content-type→hook table): scroll-stopping first line matched tohook_type, concrete evidence (a number/workflow from the source), mirror-question or either/or close, criticize systems not people, no "follow/save" CTA. - Length follows the angle — no upper limit. Write as long as the point genuinely needs to land; develop the idea fully (concrete examples, the backstage detail, the "so what") rather than truncating. Stay focused on the ONE angle and cut filler — but never artificially keep it short. "Short" here means single-angle and derived from a long source, not low word-count.
- Preserve frontmatter; keep
angle/hook_type/sourceaccurate.
- The user hits 초안 열기 / ⟳ in the browser to see your edit. Take turns: they save before handing to you; they reload after you edit.
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 b872860a51e2
- 2mo ago First seen · 48 lines · 157 tokens per session scan A 8e57b18e5cc0
shortwrite is a skill published in the GitHub repository slowww-ai/second-brain (10 stars, last pushed 1mo ago), licensed MIT. It adds 157 tokens to every session and 950 once invoked, about $0.0006 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 2026-08-31.
shortwrite compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| shortwriteslowww-ai/second-brain | Skill | 10 | 1mo ago | MIT | A | 157 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |