Borrowing it
Nothing to install: this file belongs to ianlkl11234s/mini-taiwan-pulse. 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/ianlkl11234s/mini-taiwan-pulse/develop/.claude/skills/blog-writing/SKILL.mdgit clone --depth 1 https://github.com/ianlkl11234s/mini-taiwan-pulseWrote 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/ianlkl11234s/mini-taiwan-pulse/blog-writing)<a href="https://agentmods.dev/skills/ianlkl11234s/mini-taiwan-pulse/blog-writing"><img src="https://agentmods.dev/badge/skills/ianlkl11234s/mini-taiwan-pulse/blog-writing/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/ianlkl11234s/mini-taiwan-pulse/blog-writing"><img src="https://agentmods.dev/badge/skills/ianlkl11234s/mini-taiwan-pulse/blog-writing.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 | $0.00169 | $0.02590 |
| Opus 5.5 | $0.00068 | $0.01036 |
| Sonnet 5.5 | $0.00034 | $0.00518 |
| Haiku 4.5 | $0.00017 | $0.00259 |
Grade A, and why
blog-writing scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- 回放圖層要帶 `date=`,而且那天要有快照。正式站目前有 rail、ships 的 2026-08-06(先 `ls public/embed-snapshots/<layer>/` 並 `curl -sI` 正式站確認 200)。沒有快照就不要嵌回放圖層,改嵌靜態圖層或用截圖。 How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pulse 筆記寫作流程
正本文件:
- 架構、front matter、語法、嵌入地圖限制 →
docs/features/blog/README.md - 要寫哪些、進度 →
docs/features/blog/article-plan.md
本檔只寫流程與踩過的坑,不重抄上面兩份。
1. 選題
- 讀
article-plan.md,依「建議順序」挑一篇(使用者有指定就照指定)。 - 主題只有三類,依「文章在做什麼」分(
src/blog/topics.ts):layers圖層(把資料畫成地圖)、analysis分析與模擬(用資料問問題)、making製作筆記(怎麼做出來的)。不要為單一資料種類新增主題;真的三類都放不下,先問使用者。 - 用一則訊息跟使用者確認:標題方向、主題、要不要嵌入地圖。附上建議值,讓使用者一句「照建議」就能帶過。
2. 開分支
從最新 origin/develop 開 worktree,不要在別人的分支上做:
git fetch origin
git worktree add -b docs/blog-<slug> .worktrees/blog-<slug> origin/develop
ln -s ../../node_modules .worktrees/blog-<slug>/node_modules
之後的指令一律用絕對路徑或 git -C,不要 cd 進 worktree。
3. 抽素材
-
Threads 貼文:
../knowledge-radar/threads/data/posts/*.jsonl(以id去重;只有部分貼文有text)。依計劃表的日期範圍抽出原文,存到 scratchpad,不要整包貼進對話:python3 - <<'PY' import json, glob seen, rows = set(), [] for f in glob.glob("../knowledge-radar/threads/data/posts/*.jsonl"): for line in open(f): p = json.loads(line) if p["id"] in seen or not p.get("text"): continue seen.add(p["id"]); rows.append(p) rows.sort(key=lambda p: p["timestamp"]) for p in rows: if "2026-02-13" <= p["timestamp"][:10] <= "2026-02-14": # ← 改成計劃表的日期 print(f'## {p["timestamp"][:10]} · {p["permalink"]}\n{p["text"]}\n') PY -
計劃表的素材範圍晚於最後同步日(2026-08-28)時,先用
threads-apiskill 增量同步。 -
作者口吻:
../knowledge-radar/threads/STYLE.md(第一人稱、轉折用「但」、技術細節具體到名詞和數字、不用 emoji)。長文可以有正常的句號與段落。
4. 查證數字(不可省)
- 文章裡每個數字、日期、名稱都要有出處:repo 文件、git log、資料檔筆數、程式常數。貼文說的數字也要回頭對一次。
- 查不到的不要編,也不要「合理推估」。草稿裡寫
〔待確認:…〕,交給使用者確認後才能發布。 - 同一件事出現不同數字時(例:台鐵 907/928/992 班),寫明口徑與時間點,或乾脆不用。
- 查證量大時派 subagent,要求回報「數字 → 出處檔案:行號/commit」對照表。
5. 寫 Markdown
檔案:content/blog/<slug>.md,front matter 欄位見 README。
- 開頭兩三句就回答「這是什麼、在哪裡看」,AI 摘要最常引用開頭。
- 內文 1,800–3,000 個中文字,分 4–6 個
##小節(自動進左側目錄)。 - 最後一節前放
## 常見問題,3 題,每題用### 問題加一段答案(自動產生 FAQ 結構化資料)。 - 標題含搜尋關鍵字(計劃表有列),32 字內;
description120 字內。 - 不出現內部名稱:程式識別字、資料表名、檔名、圖層 key。
- 文末不要自己放 Threads 連結,版型會依
threads:自動列出。 tags不要重複主題名。
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 Changed · +2 lines 99d14a8cd567
- 13d ago Changed 60afda85bfe3
- 14d ago First seen · 126 lines · 169 tokens per session scan A 602ceceb3946
blog-writing is a skill published in the GitHub repository ianlkl11234s/mini-taiwan-pulse (543 stars, last pushed yesterday), licensed MIT. It adds 169 tokens to every session and 2,590 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-24.
blog-writing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| blog-writingianlkl11234s/mini-taiwan-pulse | Skill | 543 | yesterday | MIT | A | 169 | 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 | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |