Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

🧬 Taiwan.md — Routine Sync（daily）v1.0 — measured AI agent skill | DecimalAI

frank890417/twmd-routine-sync

SkillScore73Maintained

讓這台機器的 routine prompt 與排程設定跟 git 的 routine SSOT 對齊。 Routine twmd-routine-sync fires 05:30 daily（晨鏈之前）; manual via "/twmd-routine-sync" or "同步 routine" or "routine 對齊". TRIGGER when: routine twmd-routine-sync fires / user says "同步 routine" / "routine 三層對賬" / 剛 ship 完 routine 改動要讓其他機器跟上.

agent-toolingby frank890417

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

DevOps & SRE pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/frank890417-twmd-routine-sync/SKILL.md

decimalai skills pull frank890417-twmd-routine-sync --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/frank890417-twmd-routine-sync/SKILL.md https://app.decimal.ai/s/frank890417-twmd-routine-sync/SKILL.md

Raw file: /s/frank890417-twmd-routine-sync/SKILL.md ↗ · machine-readable summary at /s/frank890417-twmd-routine-sync.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 2h ago

—-24% tokens0% turns73% of whole cases pass with skill (16/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-05	✗→✓	▲ Improved	-44%	0%
case-06	✗→✓	▲ Improved	1%	0%
case-07	✗→✓	▲ Improved	-61%	0%
case-08	✗→✓	▲ Improved	-37%	0%
case-10	✗→✓	▲ Improved	-38%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

frank890417 / taiwan-md

.claude/skills/twmd-routine-sync/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1

Skill contents

SKILL.md

🧬 Taiwan.md — Routine Sync（daily）v1.0

canonical 在 ROUTINE.md 註 ¹⁸。本 skill 是薄殼，只 pointer + HARD gate。

🚨 STRICT BECOME GATE

跑 /twmd-become micro 走完 Step 0-9，self-test 過才進第 1 步。

執行

bash
git pull origin main
python3 scripts/tools/routine-sync.py            # 對賬，不動檔案

exit 0 → 安靜收工。有漂移 → 先判方向再動手：git 新用 --apply --stamp $(date +%F)（機器版先存證再覆蓋），機器新用 --harvest，判不出來停手交觀察者。cron / enabled 的 live 值工具不改，用 mcp__scheduled-tasks__update_scheduled_task。

HARD gate

🔴 判不出漂移方向不猜，兩份都留著寫 Handoff
🔴 不碰非 twmd-* / taiwanmd-* 的 task（Muse 的不是我的）
🔴 SSOT 說開但 live 關、且 ROUTINE.md 註明是哲宇 directive 停用 → live 才對，不擅自打開
🔴 只 git add 自己碰的路徑，禁 git add .

收官

/twmd-finale：memory 一行寫對賬結果（幾項一致／修了什麼／方向）+ 未解漂移 + Handoff。零漂移也要記，否則看不出這條有沒有在跑。

Want to see if frank890417/twmd-routine-sync would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.