Spraay Trust & Safety
Verify before you pay. Free address safety screening (phishing, sanctions, exploits, mixers), free token honeypot checks, free transaction decoding, plus paid multi-dimensional age…
Plag
@plagtech
What This Skill Does
Screens blockchain addresses for phishing, sanctions, exploits, and mixer usage; checks tokens for honeypot risks, sell taxes, and blacklists; decodes EVM transactions into plain English; and provides multi-dimensional agent trust scores via ProofLayer. All operations go through a single x402 gateway with free safety endpoints and a paid trust score endpoint.
Replaces manual address vetting and multiple separate blockchain explorer lookups by combining safety screening, token checks, transaction decoding, and trust scoring in one unified tool.
When to Use It
- Screen a wallet address for phishing or sanctions before sending funds
- Check if a token is a honeypot before trading on Base
- Decode an EVM transaction to verify what it actually did
- Look up an agent's trust score before making a payment
- Vet a counterparty's reputation before entering an escrow agreement
- Audit a transaction hash for swaps, approvals, or batch transfers
Install
$ openclaw skills install @plagtech/spraay-trustSpraay Trust & Safety 💧
Verify before you pay. Free safety screening plus ProofLayer trust scoring for agent-to-agent commerce, via the Spraay x402 gateway.
All requests in this skill go exclusively to the user's configured
SPRAAY_GATEWAY_URL. No data is sent to any other external endpoint.
Important Notes
A trust score is a signal, not a guarantee. Scores are derived from
on-chain signals, reliability history, and reputation data. A high score
reduces risk; it does not eliminate it. For meaningful amounts, combine a
trust check with escrow (see the spraay-escrow skill) rather than relying
on either alone.
New agents score low by default. An unscored or low-scored agent is not necessarily malicious — it may simply be new. Treat "no history" and "bad history" differently when advising the user.
Setup
The gateway URL must be set in your environment or openclaw.json:
SPRAAY_GATEWAY_URL=https://gateway.spraay.app
No API key is needed. The paid trust score endpoint uses the x402 HTTP payment protocol (HTTP 402 → pay → retry). An x402-compatible wallet (Coinbase CDP or similar) handles this automatically. The three safety endpoints are completely free.
Workflows
Address Safety Screen — FREE
Screen any recipient before sending funds: phishing, sanctions, exploit involvement, mixer usage, malicious contracts.
curl "$SPRAAY_GATEWAY_URL/api/v1/address/safety?address=0xABC...123"
Run this before every first-time payment. It costs nothing.
Token Safety Check — FREE
Pre-trade token check: honeypot detection, sell tax, mint/blacklist functions, proxy risk. GoPlus-powered with Spraay severity scoring.
curl "$SPRAAY_GATEWAY_URL/api/v1/token/safety?address=0xTOKEN...&chain=base"
Decode a Transaction — FREE
Plain-English summary plus structured token transfers for any EVM transaction — swaps, transfers, approvals, wraps, NFTs, batch payments. Use it to verify what a transaction actually did.
curl "$SPRAAY_GATEWAY_URL/api/v1/tx/decode?hash=0xTXHASH...&chain=base"
Agent Trust Score — $0.03
Multi-dimensional wallet/agent trust score via ProofLayer: financial, reliability, trust, and social axes, plus XMTP reputation and on-chain signals. Counterparty due diligence for agent-to-agent payments.
curl "$SPRAAY_GATEWAY_URL/api/v1/trust/score?address=0xABC...123"
If you receive HTTP 402, the response body contains payment instructions. Pay
the facilitator, then retry with the X-PAYMENT header containing the proof.
Interpreting Results
- Safety screen fails (sanctions, phishing, exploits) — recommend against transacting, full stop.
- Safety clean + strong trust score — established counterparty. Direct payment is reasonable for routine amounts.
- Safety clean + weak or moderate score — some history. Use escrow for anything non-trivial.
- Safety clean + no score / no history — new agent. Not a red flag by itself; use escrow and small amounts first.
Recommended Pre-Payment Flow
- Run the free address safety screen. Hard stop on sanctions/phishing hits.
- For agent counterparties or larger amounts, fetch the trust score ($0.03).
- Based on results: pay directly (
spraay-payments), escrow first (spraay-escrow), or decline. - After payment, decode the transaction (free) to confirm what happened on-chain.
Free Endpoints
These require no x402 payment:
GET /api/v1/address/safety— Address safety screenGET /api/v1/token/safety— Token honeypot/risk checkGET /api/v1/tx/decode— Transaction decoderGET /health— Gateway health check
x402 Payment Flow
- Call the trust score endpoint.
- Receive HTTP 402 with payment details (
acceptsarray with amount, asset, and payment address). - Agent wallet sends the micropayment ($0.03 in USDC on Base).
- Retry the request with the
X-PAYMENTproof header. - Receive the response.
Error Handling
402— Payment required. Follow instructions in response body.400— Bad request. Check the address/hash format.404— No record for this address (new agent — see interpretation notes).500— Server error. Retry after a moment.
Tips
- The safety screen is free — run it before every first payment, always.
- Reserve the paid trust score for agent counterparties and larger amounts.
- Pair a weak/no-score result with
spraay-escrowinstead of refusing outright. - Decode transactions after the fact to verify delivery of on-chain work.
Links
- App: https://spraay.app
- ProofLayer: https://prooflayer.net
- Docs: https://docs.spraay.app
- GitHub: https://github.com/plagtech
Top skills in this category
Skill Vetter
@spclaudehomeSecurity-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
Github
@steipeteInteract with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
Humanizer
@biostartechnologyRemove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.
Free Ride - Unlimited free AI
@shaivpidadiManages free AI models from OpenRouter for OpenClaw. Automatically ranks models by quality, configures fallbacks for rate-limit handling, and updates opencla...
Elite Longterm Memory
@nextfrontierbuildsUltimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.