PlainText Wallet Approval Checker
RemoteAbout PlainText Wallet Approval Checker
PlainText is a Node.js app for people who need to understand token and NFT approvals or signature requests. It explains checks in plain language and labels results SAFE, CAUTION, or RISK; verdicts are currently based on presign-guard.
Key features
- 01Check token and NFT approvals
- 02Explain check results in plain language
- 03Label results SAFE, CAUTION, or RISK
- 04Process checks statelessly without storing user data
- 05Accept feedback submissions
Use cases
- 01Review token or NFT approvals for a wallet
- 02Understand an approval or transaction before signing
- 03Submit bug reports or feature suggestions
MCP support
PlainText offers an MCP feedback tool for clients to submit bugs, feature requests, or other feedback. The README also describes using presign-guard for verdicts and reason codes.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://plaintext.fizzl.eu/mcpclaude mcp add --transport http fizzl13-plaintext-wallet-checker https://plaintext.fizzl.eu/mcp{"mcpServers":{"fizzl13-plaintext-wallet-checker":{"type":"http","url":"https://plaintext.fizzl.eu/mcp"}}}Open in Cursor{"servers":{"fizzl13-plaintext-wallet-checker":{"type":"http","url":"https://plaintext.fizzl.eu/mcp"}}}Open in VS Code{"mcpServers":{"fizzl13-plaintext-wallet-checker":{"type":"http","url":"https://plaintext.fizzl.eu/mcp"}}}{"mcpServers":{"fizzl13-plaintext-wallet-checker":{"serverUrl":"https://plaintext.fizzl.eu/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEPlainText — wallet approval translator
Built on presign-guard: PlainText's verdicts now come from presign-guard (the same checks, sanctions list, wallet age and signed receipt); PlainText adds the explanation in plain language for people.
POST /api/check-wallet($0.04) uses presign-guard's approval audit,POST /api/explain($0.03) its pre-sign check when the payload is a real approval, transaction or typed-data signature (anything else is explained by the model alone,source: "model"). Agents that only need the verdict and reason codes can call presign-guard directly ($0.01–$0.02). NeedsFIZZL_INTERNAL_KEY(the same value on both services, Render only); without it the old path (GoPlus + Claude) runs.
A small Node.js app that looks up a wallet's token/NFT approvals via the GoPlus Security API and asks Claude to explain them in plain language.
▶ Watch the 1-minute explainer · Live: https://plaintext.fizzl.eu
Project structure
plaintext-wallet/
├── server.js Express backend (GoPlus + Anthropic calls)
├── media.js Serves the explainer video from the explainer-video branch
├── feedback.js POST /feedback and the MCP tool feedback (same file in every Fizzl service)
├── media/explainer/ Script and pipeline for the explainer video
├── package.json
├── render.yaml Render deploy config (optional, see below)
├── .env.example Copy to .env for local dev
└── public/
└── index.html Frontend (wallet check + demo scenarios)
Feedback
Found a bug, or missing something? Send it with POST /feedback (free, no payment) or the MCP tool feedback:
curl -X POST https://plaintext.fizzl.eu/feedback \
-H 'content-type: application/json' \
-d '{"type": "feature", "message": "Support Solana wallets", "endpoint": "/api/check-wallet"}'
type is bug, feature or other; message is required (up to 2000 characters); endpoint and contact are optional. The answer is 202 with an id. At most 10 reports per hour per caller. Reports go to the usage log, and a person reads every one; nothing in a report is run or changed automatically. GET /feedback shows the schema.
1. Run it locally first
cd plaintext-wallet
npm install
cp .env.example .env
# edit .env and paste your ANTHROPIC_API_KEY
npm start
Open http://localhost:3000 — try the "Try a sample" tab first (no API keys needed besides Anthropic), then "Check a wallet" with a real address once you're happy it works.
2. Deploy to Render
Option A — one-click via render.yaml (Blueprint)
- Push this folder to a new GitHub repo.
- In the Render dashboard: New + → Blueprint → select your repo.
Render reads
render.yamlautomatically and creates the web service. - When prompted, fill in the environment variables:
ANTHROPIC_API_KEY— required, from console.anthropic.comGOPLUS_ACCESS_TOKEN— optional, only needed if you outgrow GoPlus's free 30 calls/minute limit (see console.gopluslabs.io)
- Click Apply — Render builds and deploys automatically.
Option B — manual web service (no render.yaml needed)
- Push this folder to a GitHub repo.
- Render dashboard → New + → Web Service → connect your repo.
- Settings:
- Environment: Node
- Build Command:
npm install - Start Command:
npm start - Instance Type: Free (fine for a prototype)
- Under Environment Variables, add
ANTHROPIC_API_KEY(and optionallyGOPLUS_ACCESS_TOKEN). - Click Create Web Service. Render gives you a
https://<name>.onrender.comURL once the build finishes.
Render's free tier spins the service down after inactivity — the first request after idling will take ~30-50 seconds to wake up. That's normal and fine for a prototype you're sharing for feedback.
Notes on the GoPlus integration
- The wallet check calls
token_approval_securityornft721_approval_securityunderhttps://api.gopluslabs.io/api/v2/.... This is documented but the GoPlus docs site is JS-rendered, so before you rely on this in front of real users, do one manual test call (e.g. withcurl) against a wallet you control, and compare the actual JSON shape to whatserver.jsexpects — field names occasionally shift between GoPlus API versions. - The free GoPlus tier is capped at 30 calls/minute — plenty for early testing, but rate-limit errors will surface as a plain error message in the UI if you exceed it.
- No user data is stored anywhere in this app — every check is a stateless round trip. If you later add accounts, saved wallets, or history, update this note (and your own privacy policy) accordingly.
Costs to expect
- Render free tier: $0/month, with the cold-start behavior above. A paid tier ($7/month "Starter") removes the spin-down if this gets real usage.
- Anthropic API: pay-per-use, billed to whatever key you provide. A single explanation call is small (well under 1K tokens), so cost per check will be a fraction of a cent — but keep an eye on your usage dashboard if this gets shared widely, since there's no rate limiting built into this prototype yet.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.
