Uniswap Seek Protocol Fees

Analyze TokenJar profitability and optionally execute a Firepit burn-and-claim. Autonomous pipeline: checks balances, prices assets, calculates profit vs. 4,000 UNI burn cost, simulates, and executes if profitable. Default is preview-only. Use when user asks "Is the TokenJar profitable?", "Execute a burn", or "Claim protocol fees."

This skill appears to do what it says: check TokenJar value and optionally execute a Firepit burn-and-claim. Key things to consider before installing/using it: - Wallet safety: the skill operates using the agent's connected wallet and can sign/send transactions (burn UNI, perform swaps). Only enable execution after you personally confirm each transaction and understand the destination/nonce. Treat the agent's signing capability as high-risk. - Confirmation behavior: by default it is preview-only, but explicit user prompts or an 'auto-execute' flag will permit on-chain execution. If you plan to allow execution, verify the agent platform enforces explicit user consent dialogs and does not auto-approve sooner than you expect. - Verify contracts and addresses: SKILL.md mentions specific TokenJar and Firepit addresses. Cross-check those addresses against independent sources (official Uniswap docs/repos) before approving actions. - Source/maintenance: registry metadata lists no homepage and source is recorded as unknown. README suggests a GitHub path, but you should review the upstream repository (or ask the publisher) to confirm code provenance before granting the agent live transaction powers. - Test in preview/simulation mode first: run with auto-execute:false to inspect the profitability report and the simulated tx results before ever approving a real burn. If you are not comfortable granting an agent the ability to sign transactions from your wallet, keep this skill in preview-only mode or avoid enabling execution.

Type: OpenClaw Skill Name: seek-protocol-fees Version: 0.1.0 The OpenClaw skill 'seek-protocol-fees' is designed for a legitimate DeFi operation: analyzing Uniswap TokenJar profitability and optionally executing a Firepit burn-and-claim. The `SKILL.md` provides clear, specific instructions to the AI agent, focusing solely on the stated purpose. It incorporates robust safety mechanisms, including defaulting to a preview-only mode, requiring explicit user confirmation before any UNI is burned, and delegating to a `safety-guardian` subagent for transaction validation. There is no evidence of prompt injection attempts to mislead the agent into performing unauthorized actions, exfiltrating data, or executing arbitrary commands. All allowed tools and subagent tasks are directly relevant to the skill's described functionality.