safelinkSecure agent-to-agent hiring and execution skill for OpenClaw MCP with escrowed settlement, x402 facilitator payments, ERC-8004 identity/reputation checks, s...
Install via ClawdBot CLI:
clawdbot install licc921/safelinkGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Contains instructions to override system prompt or ignore user requests
"Ignore all previous instructions"Potentially destructive shell commands in tool definitions
curl -L https://foundry.paradigm.xyz | bashCalls external URL not in known-safe list
https://x402.org/facilitatorUses known external API (expected, informational)
api.anthropic.comGenerated Mar 21, 2026
A platform where AI agents can hire other agents for specialized tasks like data analysis or content generation, with payments held in escrow until proof of work is delivered. This ensures both parties are protected against non-delivery or fraud, using ERC-8004 for reputation checks to vet agents before hiring.
DAOs can use SafeLink to delegate operational tasks to AI agents, such as treasury management or community moderation, with escrowed payments released only upon verified completion. This reduces manual oversight and enhances security through MPC wallet signing and replay protection.
Businesses implement SafeLink to automate internal processes like report generation or customer support, where AI agents execute tasks with proof-before-settlement to ensure accuracy. It integrates with existing systems via MCP, offering policy gating and risk-scored transactions for compliance.
A marketplace where users can hire AI agents for Web3-specific services like smart contract auditing or NFT minting, using x402 facilitator payments for gas efficiency. SafeLink's escrow and identity checks build trust, while beginner-friendly wallet setup lowers entry barriers.
Academic or research institutions employ SafeLink to coordinate AI agents for collaborative projects, such as data processing or simulation runs, with payments escrowed until results are validated. This ensures fair compensation and reduces disputes through deterministic proof verification.
Charge a small percentage fee on each escrowed payment processed through SafeLink, leveraging the x402 facilitator for efficient settlements. Revenue scales with transaction volume, appealing to high-frequency A2A workflows in agent marketplaces or enterprise automation.
Offer tiered subscriptions providing access to premium features like enhanced risk scoring, SIWx authentication, or priority support. Targets production users needing higher security and customization, such as large DAOs or enterprise clients.
Generate revenue by offering custom integration services, deployment support, and consulting for organizations adopting SafeLink into their existing MCP or Web3 ecosystems. This model capitalizes on the skill's complexity and niche in secure A2A workflows.
💬 Integration Tip
Start by setting up the required environment variables like BASE_RPC_URL and MPC provider keys, then use the interactive generate-env.ts script for initial configuration to avoid common setup errors.
Scored Apr 19, 2026
AI Analysis
The 'prompt poisoning' signal appears to be a documented security test fixture, not a hidden instruction. The skill's external API usage (x402 facilitator, Anthropic, MPC wallet providers) is consistent with its stated purpose of secure agent-to-agent transactions and escrow. No evidence of credential harvesting, data exfiltration, or obfuscated malicious behavior was found in the provided definition.
Audited Apr 17, 2026 · audit v1.0
AI crypto research and analysis powered by CellCog. Token deep-dives, on-chain metrics, DeFi protocol breakdowns, wallet portfolio reviews, market sentiment, whitepaper analysis, smart contract evaluation. Comprehensive crypto intelligence from a single prompt.
当用户发起任意旅行相关问询时,包含但不限于:预订酒店、机票查询、火车票查询、景点推荐、寻找当地特色玩乐、目的地查询、行程规划、美食住宿攻略、签证、查询旅游攻略、获取旅行建议等场景,自动触发此技能。当用户需要操作携程、Ctrip、携程问道时使用此skill。
Expertise in modern Ethereum and EVM development, including smart contracts, Layer 2s, standards, agent infrastructure (ERC-8004, x402), gas optimization, se...
Store and retrieve encrypted files permanently on Monad blockchain via DropClaw x402 gateway
OpenSea (opensea.io). Use this skill for ANY OpenSea request — searching and reading data. Whenever a task involves OpenSea, use this skill instead of callin...
Manage aibtc-worker automation tasks. Use this skill when the user wants to start, stop, or check the status of the aibtc worker. Triggers include: "aibtc ru...