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 YuDefine/nuxt-supabase-starter --skill blog-scoutgit clone --depth 1 https://github.com/YuDefine/nuxt-supabase-starterWrote 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/yudefine/nuxt-supabase-starter/blog-scout)<a href="https://agentmods.dev/skills/yudefine/nuxt-supabase-starter/blog-scout"><img src="https://agentmods.dev/badge/skills/yudefine/nuxt-supabase-starter/blog-scout/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/yudefine/nuxt-supabase-starter/blog-scout"><img src="https://agentmods.dev/badge/skills/yudefine/nuxt-supabase-starter/blog-scout.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 | $0.00158 | $0.03307 |
| Opus 5 | $0.00079 | $0.01654 |
| Sonnet 5 | $0.00032 | $0.00661 |
| Haiku 4.5 | $0.00016 | $0.00331 |
Grade A, and why
blog-scout scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
blog-scout — 內部知識資產 → 部落格選題建議
把「哪些坑夠通用、哪些已寫過、哪些湊在一起才成一篇、哪些敏感度太高不能寫」這四個人腦判斷標準化成可重跑的掃描流程。輸入是 clade 的知識資產現況，輸出是一份排序過的選題建議報告——到此為止。
兩條硬邊界（先讀）
- 只選題，NEVER 寫成稿。文章成稿一律主線 Opus 自己寫（
rules/core/agent-routing.md§ 派不派：對外文件的定稿措辭外包不了）。本 skill 的輸出止於「建議題目 + 素材清單 + 敏感度標記」。 - NEVER 寫入 。本 skill 對
~/offline/<consumer-k>/只讀不寫；建議被採納後的落地（開稿、frontmatter、category）由使用者與主線另行處理。
判準來源（Step 1 必讀，NEVER 信本檔的轉述）
| 判準 | SoT | 用在哪 |
|---|---|---|
| 寫什麼 / 不寫什麼、藏招準則、禁詞清單 | ~/offline/<consumer-k>/CLAUDE.md § 部落格內容主軸 |
敏感度分級、候選准入 |
| 讀者輪廓與品牌定位 | ~/offline/<consumer-k>/PRODUCT.md |
每題「讀者是誰 / 為什麼值得讀」 |
| pitfall frontmatter 契約與合格標準 | ~/offline/clade/docs/pitfalls/README.md |
素材解析 |
| tag controlled vocabulary | ~/offline/clade/docs/pitfalls/tags.yml |
聚類與內部 tag 判定 |
禁詞清單與內部 tag 清單在 scripts/scan-pitfalls.py 有內建副本，會漂移：Step 1 讀完 SoT 後發現新禁詞，用 --banned-extra 補進掃描，NEVER 直接信 script 內建清單。
Workflow
Step 1 — 載入判準
讀上表四個 SoT（各只讀相關節）。輸出：本輪生效的禁詞清單、藏招準則三欄表、現行 category 清單。
Step 2 — 素材盤點與聚類（維度 1）
python3 "$SKILL_DIR/scripts/scan-pitfalls.py" --min-cluster 5
（$SKILL_DIR = 本 skill 的 base directory，skill 載入時 harness 會告知；archive 也要納入時加 --include-archive。）
輸出三段：inventory（條目數 / severity / status / 年月分佈）、clusters（public-tag pair 聚類，每群列成員 + severity + 敏感度初判）、sensitivity 統計。實跑形狀（2026-08-25，236 條）：
severity: critical=8 high=140 mid=83 low=5
### cli-tooling + silent-failure — 24 條（低=0 中=8 高=16）
- [critical/中] 2026-05-14-publish-flow-cleans-parallel-untracked.md — 並行 …
低=2 中=82 高=152
聚類是按「可寫成員數」（低+中）排，不是按原始條數。 這兩者在本資料集上幾乎反相關——
最大的幾群（cli-tooling / git / cross-session）全是內部工具坑，可寫成員為 0。
所以排在前面的不一定是最大的群，這是刻意的；不要以為輸出壞了。
另外注意 低 這一級實質上是空的（236 條裡只有 2 條）：機械初判把 body 含任何禁詞的都打成
中，而幾乎每條 pitfall 都會提到某個內部專案名。中 才是素材的主體，不要把它當成次級品。
聚類是選題單位，不是逐條列出：一篇文章的理想素材是同群 3–5 條「低/中」敏感度成員（既有文章型態的實證：已發表的多坑文都是 4–5 坑一篇）。群內「高」敏感成員直接剔除，不因它們放棄整群。
Step 3 — 第二素材源：跨 consumer 最佳實踐比較（維度 4）
jq -r '.conventions[] | select((.variants|length)>=2) |
"\(.convention_id) [\(.layer)] " + ([.variants[] | .variant_id + "(" + .maturity +
(if .migration_to then "→"+.migration_to else "" end) + ")"] | join(" / "))' \
~/offline/clade/registry/conventions.json
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today Changed · +6 lines 28db6791806f
- yesterday Changed · -2 tokens per session 44a1f36a3a85
- 7d ago First seen · 155 lines · 160 tokens per session scan A 750a9fbe4165
blog-scout is a skill published in the GitHub repository YuDefine/nuxt-supabase-starter (45 stars, last pushed today), licensed MIT. It adds 158 tokens to every session and 3,307 once invoked, about $0.0008 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 2026-09-03.
Other skills, from other repositories
prd-v07-implementation-loop
Execute implementation within EPICs following test-first development, continuous SoT updates, and code traceability during PRD v0.7 Build Execution. Triggers on requests to start building, implement an epic, begin coding, or when user asks "start building", "implement epic", "coding", "development", "build execution"…
prd-v07-test-planning
Define test cases BEFORE implementation, ensuring every API, business rule, and user journey has verifiable acceptance criteria during PRD v0.7 Build Execution. Triggers on requests to define tests, plan test coverage, create test cases, or when user asks "define tests", "test planning", "what to test?", "test cases"…
strict-tdd
Strict RED->GREEN->REFACTOR test-driven development with enforcement. Never write production code before a failing test. Atomic commits per TDD cycle.
prd-v04-user-journey-mapping
Map user missions from trigger to value moment, organizing features into coherent paths during PRD v0.4 User Journeys. Triggers on requests to map user journeys, define user flows, describe how users accomplish goals, or when user asks "map user journeys", "define user flows", "user missions", "how do users accomplish…
atomic-tdd
Test-first discipline. Auto-triggers on "let's implement X", "add feature Y", "fix bug Z", "write a test for", "implement", "build out", and similar pre-code-change phrases. Iron rule: failing test exists before production code. Skip only for pure docs/config changes with an explicit "skipped because:" note. Explicit…
use-tdd
Implement a requested increment with red-green-refactor. Use when the user types /use-tdd.