Uniswap Deploy Agent Token
SuspiciousAudited by ClawScan on May 10, 2026.
Overview
This skill is coherent for launching a Uniswap token, but it delegates irreversible on-chain financial actions to unreviewed agents without declaring wallet permissions or explicit approval limits.
Review carefully before installing. Only use this with a wallet that requires manual signing, confirm every transaction and contract address yourself, set strict spend/slippage limits, and do not rely on the referenced subagents unless their code and permissions are available for review.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A mistaken or unsafe invocation could spend funds, create a public pool with bad parameters, or lock liquidity in a way the user cannot easily undo.
The skill directs a delegated agent to perform irreversible on-chain financial actions, but the visible workflow does not require a final user review of transactions, spending limits, slippage, recipient contracts, or lock terms before execution.
The agent executes a 7-step pipeline: ... **Create pool** ... execute initialization. ... **Bootstrap liquidity** ... add liquidity via position manager. ... **Lock LP**: Transfer position NFT to time-locked vault
Require explicit user confirmation before every on-chain transaction, show exact calldata/contract addresses/amounts/fees/lock duration, and enforce user-specified spending and slippage limits.
Users are not clearly told what wallet/account authority the skill needs or how that authority is constrained before it can affect crypto assets.
The registry declares no credential contract, while the SKILL.md workflow implies wallet or signing authority to initialize pools, add liquidity, and transfer LP NFTs.
Required env vars: none ... Primary credential: none
Declare the required wallet/signing mechanism, supported chains, permission scope, and approval model; avoid granting automated transaction authority beyond user-reviewed signing.
The actual logic that would prepare or execute financial transactions may come from components the user has not inspected or installed knowingly.
The high-impact deployment pipeline depends on subagents whose implementations are not present in the reviewed file manifest, leaving their provenance, limits, and safety checks unverified.
allowed-tools: [Task(subagent_type:token-deployer)] ... The agent internally delegates to `safety-guardian` ... and `lp-strategist`
Include or pin the referenced subagent definitions, document their permissions, and make the skill safe to use even if delegated agents are unavailable or untrusted.