Deploy Spark Bitcoin L2 Proxy

Deploy a serverless Spark Bitcoin L2 proxy on Vercel with spending limits, auth, and Redis logging. Use when user wants to set up a new proxy, configure env...

This skill does what it says (deploys a wallet-backed proxy) but it requires handing over highly sensitive secrets — most importantly the wallet mnemonic and cloud/service tokens — and the registry metadata did not list those requirements. Before proceeding: 1) Do not paste your real mnemonic or long-term keys into anything you haven't audited; consider using a throwaway/test mnemonic with minimal funds for initial testing. 2) Review the GitHub repository and its package.json (npm deps) for unexpected code or network calls before running npm install. 3) Limit Upstash and Vercel credential scopes where possible, create least-privilege API keys, and rotate them after setup. 4) Use the proxy's scoped tokens (invoice/admin) and set conservative MAX_TRANSACTION_SATS and DAILY_BUDGET_SATS values. 5) Ask the publisher to fix the registry metadata to enumerate the required env vars and describe exactly what each secret is used for. If you are uncomfortable storing a mnemonic in Vercel envs, consider self-hosting or alternative architectures (read-only endpoints, custodial services, or hardware-based signing).

Type: OpenClaw Skill Name: sparkbtcbot-proxy-deploy Version: 1.0.0 The skill instructs the AI agent to perform high-risk operations necessary for deploying a serverless proxy, including cloning and executing code from an external GitHub repository (`echennells/sparkbtcbot-proxy.git`), making network calls with user-provided credentials (Upstash API key, Vercel token), and handling sensitive data like a BIP39 mnemonic. While these actions are aligned with the stated purpose of deploying the proxy, the extensive use of shell commands with interpolated values (e.g., `curl` commands in SKILL.md) creates a significant shell injection vulnerability if the AI agent does not rigorously sanitize user inputs. There is no evidence of intentional malicious behavior such as data exfiltration or backdoor installation within the skill's instructions, but the inherent risks associated with these operations warrant a 'suspicious' classification.