What does this skill do, and when should you use it?
PinMe is a zero-configuration deployment tool that allows uploading static files to IPFS or creating and deploying full-stack projects with frontend (React+Vite), backend (Cloudflare Worker), and database (D1 SQLite). This skill provides detailed command flows, project structures, and code patterns for agents to use when handling 'pinme' requests or when backend support is needed. It requires authentication via `pinme login` or AppKey, and depends on the npm-installed `pinme` CLI.
The skill guides agents to: 1) check and update the pinme CLI to the latest version; 2) authenticate via pinme login or pinme set-appkey <AppKey>; 3) for Path 1 (upload files/static sites): determine build output directory (dist/build/out/public), run pinme upload <path> to upload, and return the final URL; 4) for Path 2 (full-stack projects): run pinme create <name> to create a project, modify template code (Worker, frontend, SQL), execute pinme save for one-command deployment, or use pinme update-worker, pinme update-web, pinme update-db to update specific parts; 5) implement JSON APIs, database operations (D1), file storage (R2), and email sending (via PinMe API) in the Worker.
- A user needs to quickly deploy a static site or folder (like build artifacts) to IPFS and get a shareable preview URL.
- A user wants to scaffold a full-stack app (React frontend + Cloudflare Worker backend + D1 database) from scratch and deploy with a single command.
- A developer has an existing PinMe project and only modified backend code, using `pinme update-worker` to update the Worker without redeploying the frontend.
- A project needs database migrations and `pinme update-db` is used to push SQL migration files.
- A backend needs to send emails, leveraging the PinMe API for email integration.
How do you install this skill?
- Skill relies on external services (IPFS, Cloudflare) that may be inaccessible in mainland China; suggest clearly indicating this to users.
- Confirm user intent before uploading to avoid accidental upload of sensitive files.
- Shell / CLI
- Network access
Node.js >= 16.13.0npmpinme CLI
The skill is located in the monorepo glitternetwork/pinme. To install the whole skill collection: npx skills add glitternetwork/pinme. Ensure Node.js >= 16.13.0 and install the CLI globally: npm install -g pinme.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/glitternetwork/pinme.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/pinme" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Use pinme to deploy my frontend
- create a full-stack project
In an Agent Skills-compatible client (e.g., Claude Code), trigger it with a prompt like 'Use pinme to deploy my frontend' or 'create a full-stack project'. For static files: run pinme login (or pinme set-appkey <AppKey>), then execute pinme upload <dist|build|out|public> and return the printed URL. For full-stack: run pinme login, pinme create my-app, modify the template code, then run pinme save from the project root. Ensure you return the full preview URL including all hash characters.
What are this skill's strengths and limitations?
- Zero-configuration, single-command deployment for full-stack apps, no manual cloud resource setup.
- IPFS-based static hosting with no hosting fees, decentralized.
- Comprehensive project template (React+Vite+Worker+D1) with guidance for customization.
- Incremental update commands (update-worker, update-web, update-db) for faster deployments.
- Requires user login or AppKey, which may not suit fully automated unattended scenarios.
- Upload size limits (default 100MB for single file, 500MB for directory) and potential network dependency.
- No test suite or explicit support guarantees mentioned in docs; transparency for production use is limited.
- Platform-specific to PinMe; not applicable if users want to use other deployment platforms.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| PinMe this page | 46 · Use with care | ★ 3.7k | 3mo ago | MIT |
| Expo EAS Hosting Deployments | 48 · Use with care | ★ 2.7k | 6d ago | MIT |
| Cloudflare Wrangler Deployment Guide | 49 · Use with care | ★ 3k | 3d ago | Apache-2.0 |
| PinMe Worker OpenRouter API Integration | 64 · Recommended | ★ 3.7k | 3mo ago | MIT |
| PinMe Worker R2 Storage Skill | 63 · Recommended | ★ 3.7k | 3mo ago | MIT |
The skill implicitly compares to platform deployment tools like Netlify or Vercel, but these are not explicitly mentioned. Given its focus on IPFS and Cloudflare Workers, it may be more suitable for decentralized hosting scenarios.
How did FollowSkills review this skill?
Evidence shows clear deployment flow and 'do not upload' list, but permissions and confirmation mechanisms rely solely on CLI login, lacking detail on sensitive data handling. Deductions: least privilege not explicit, no rollback plan.
CI tests cover some command success paths, but static review cannot verify actual execution. Deductions: incomplete test coverage, insufficient error handling details.
Skill description is clear with decision diagram and non-fit ranges, but environment fit is limited, especially unclear access to mainland China networks. Deductions: insufficient evidence for boundaries and trigger conditions.
Documentation is well-structured with command reference and common errors, but lacks version history and ownership information. Deductions: maintenance responsibility and update path unclear.
Core task is achievable, but static review cannot verify direct usability of outputs. Deductions: lack of verifiable deployment results.
Test files exist but coverage is incomplete, and no independent verification evidence. Deductions: limited third-party execution evidence.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →