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 cf-cligit 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/cf-cli)<a href="https://agentmods.dev/skills/yudefine/nuxt-supabase-starter/cf-cli"><img src="https://agentmods.dev/badge/skills/yudefine/nuxt-supabase-starter/cf-cli/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/cf-cli"><img src="https://agentmods.dev/badge/skills/yudefine/nuxt-supabase-starter/cf-cli.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 | $0.00066 | $0.01185 |
| Opus 5.5 | $0.00026 | $0.00474 |
| Sonnet 5.5 | $0.00013 | $0.00237 |
| Haiku 4.5 | $0.00007 | $0.00119 |
Grade B, and why
cf-cli scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
( f="$(mktemp)"; trap 'rm -f "$f"' EXIT; trap 'exit 130' INT TERM; chmod 600 "$f" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
description: 'Use when 查詢或異動 Cloudflare 帳號資源(D1/KV/R2/DNS/zone/tunnel/Worker secret/cache/browser-run),或把 wrangler / curl 帳號操作改寫成 cf。NOT for deploy / dev / tail / types(走 wrangler)。' How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cf CLI
帳號資源操作一律走官方 cf CLI(npm cf,github.com/cloudflare/cf)。哪些操作不走 cf(deploy、dev、CI 步驟、程式碼內 API client、cf 未覆蓋端點)見 cloudflare-workers.md § 8。cf 仍是 beta,本檔只寫已實測的語意差,NEVER 把這裡或記憶中的指令當權威。
起手
- 安裝:
mise use -g npm:[email protected](NEVERnpm i -g;MUST 指定版本——beta 期mise latest npm:cf解析到舊的0.15.0,不是本檔實測的版本),cf --version確認 - 認證:
CLOUDFLARE_API_TOKEN(account 用CLOUDFLARE_ACCOUNT_ID),或cf auth login/--profile。多帳號時異動前 MUSTcf auth whoami - 找指令:第一次用某個子指令、或指令失敗時,
cf cli search "<動作 + 資源類型>"→<指令> --help;API 欄位用cf schema <指令去掉開頭 cf>。查詢字串 NEVER 帶 domain、account / zone / resource ID、email、token - 輸出:成功時 stdout 是 API 回應的
result本體(沒有.result外層),API 錯誤時 exit 非 0;jq路徑照此寫 --dry-run:異動類指令 MUST 先跑。它只在本機印出將送出的 request,不經 API 驗證。其他檔(skill、snippet、runbook)的異動範例為了簡潔省略這一步,NEVER 讀成可跳過
跟 wrangler 不同、會出事的語意
| 面向 | wrangler | cf |
|---|---|---|
| D1 指令目標 | database name | database UUID(cf d1 list --name <name> 取得) |
| D1 預設打哪裡 | 本機(遠端要 --remote) |
遠端 production |
--local |
專案 dev server 的本機 state | cf 自己的 state(~/.config/cloudflare/state),看不到 dev server 的資料——查 dev 資料用 wrangler d1 execute <db> --local,dev server 的 state 不在 wrangler 預設目錄(例如 NuxtHub dev)時加 --persist-to <該目錄> |
| D1 migrations 目錄 | wrangler config migrations_dir |
--dir(預設 ./migrations);Drizzle 巢狀 layout 加 --pattern "<dir>/*/migration.sql" |
把 wrangler d1 execute <name>(本機)直接換成 cf d1 query <uuid> 等於改打 production。改寫舊指令時逐列對照上表。
Worker secret
手動推 secret 只限 secrets.md § 唯一例外的情境。推的時候值不進 argv:
( f="$(mktemp)"; trap 'rm -f "$f"' EXIT; trap 'exit 130' INT TERM; chmod 600 "$f"
"${EDITOR:-vi}" "$f"
# {"<NAME>": {"type": "secret_text", "text": "<值>"}, ...} ← JSON merge patch,key 是 secret 名稱
cf workers secrets bulk --worker <worker-name> --file "$f" )
- 包在 subshell 裡是為了讓
trap在這段結束時就刪檔,而不是等互動 shell 關掉;Ctrl-C/TERM 會中止整段,不會帶著已刪的檔去跑cf - 編輯器會在暫存檔旁留 swap/backup/undo 副本,那些是明文:vim 用
vim -n -i NONE -c 'set nobackup nowritebackup noundofile',其他編輯器先確認同等設定
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.
- 8d ago Changed · -1 lines 3ca325954490
- 16d ago First seen · 48 lines · 66 tokens per session scan B 5131cde4ec4c
cf-cli is a skill published in the GitHub repository YuDefine/nuxt-supabase-starter (44 stars, last pushed 2d ago), licensed MIT. It adds 66 tokens to every session and 1,185 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-01.
Other skills, from other repositories
web-design
A tool for designing and publishing single-page HTML websites, such as reports, landing pages, and data displays. It uses common web libraries for layouts, charts, icons, and responsive pages, then can deploy the result to Cloudflare Pages.
nw-investigation-techniques
Evidence collection methods, problem categorization, analysis techniques, and solution design patterns.
nw-par-critique-dimensions
Platform design review critique dimensions and severity levels. Load when reviewing CI/CD pipelines, infrastructure, deployment strategies, observability, or security designs.
nw-deployment-strategies
Rollback procedures, risk assessment, pre/post-deployment validation, and contingency planning. Load when orchestrating deployment or preparing rollback plans. For deployment strategy details (canary, blue-green, rolling), see cicd-and-deployment skill.
viberaven
Use VibeRaven when working on an AI-built app that must ship to real users; when the user asks to ship, deploy, go to production, connect auth, database, payments, RLS, env vars, webhooks, or monitoring; when production-only errors appear; or when the agent needs product context (versions, providers, launch risks)…
vercel-supabase-launch-check
Use when an AI-built app on Vercel and Supabase (Lovable, Bolt, v0, Cursor or Claude Code output) is about to launch or be handed to a client, or when the user asks whether Supabase tables are missing RLS, whether a policy lets everyone read or write, whether a service role or secret key leaked into client code or a…