Back to skill

Security audit

Uniswap Deploy Agent Token

Security checks for vulnerabilities and agentic risk

Overview

The skill is a clearly described crypto-token deployment helper, but it needs review because it can trigger irreversible on-chain financial actions and its install instructions use mutable remote packages.

Review this before installing because it automates real crypto deployment actions. Use only a pinned, reviewed install source, run it with a dedicated low-privilege wallet, verify every pool, hook, liquidity, price, and LP-lock parameter before signing, and do not let it broadcast transactions without explicit confirmation.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
README.md:9
Finding
Unpinned Remote Dependencies in Installation Commands<![CDATA[ ## Vulnerability Details **File Location**: `README.md`, lines 9–19 **Vulnerability Type**: T08: Insecure Dependencies **Risk Level**: Medium ### Vulnerable Code ```bash Install into Claude Code or Cursor with: ```bash npx skills add https://github.com/wpank/Agentic-Uniswap/tree/main/.ai/skills/deploy-agent-token ``` Or via Clawhub: ```bash npx clawhub@latest install deploy-agent-token ``` ``` ### Technical Analysis The documented installation procedures rely on remote, mutable resources without pinning them to reviewed versions or verifying their integrity. The first command executes the unpinned `skills` npm CLI and installs Skill content from the mutable `main` branch of a GitHub repository. The second command explicitly executes the mutable `latest` release of the `clawhub` npm package. Neither command specifies a trusted package version, immutable Git commit, cryptographic checksum, or signature. Because `npx` downloads and executes package code, the effective code run by these commands can change after this audit. A compromised npm publisher account, malicious package update, compromised GitHub account, or modification to the repository's `main` branch could introduce attacker-controlled code or Skill instructions. No evidence was found that the currently audited files contain a malicious payload. The risk arises from the mutable, unverified installation chain. ### Attack Path 1. An attacker compromises the relevant npm publisher account, package distribution channel, GitHub repository, or maintainer credentials. 2. The attacker publishes a malicious CLI release or modifies content under the referenced GitHub `main` branch. 3. A user follows the installation instructions and executes one of the documented `npx` commands. 4. `npx` downloads and executes the modified CLI with the privileges of the invoking user. 5. The malicious installer can alter files accessible to that user, install manipulated Skill instructions, access user-read ...[truncated 827 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Pin every npm CLI to an exact reviewed version instead of relying on an implicit version or `@latest`, for example: ```bash npx skills@<reviewed-version> add ... npx clawhub@<reviewed-version> install deploy-agent-token ``` 2. Replace the mutable GitHub `main` branch reference with an immutable, reviewed commit SHA or signed release tag. 3. Publish and verify cryptographic checksums or signatures for downloaded Skill artifacts before installation. 4. Prefer installing reviewed CLI dependencies through a lockfile and then invoking the local pinned binary rather than downloading and executing packages directly through `npx`. 5. Review package lifecycle scripts and the complete dependency tree before approving new versions. 6. Run installation in a least-privileged, isolated environment without production credentials or sensitive environment variables. 7. Establish a controlled update process in which new dependency versions and upstream Skill revisions are audited before documentation is updated. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding
The README instructs users to run `npx skills add ...` without pinning an exact package version, which causes execution of whatever version is current at install time. If the upstream package is compromised, typosquatted, or publishes a malicious update, users could execute attacker-controlled code during installation.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding
The command `npx clawhub@latest install deploy-agent-token` explicitly tracks the latest published version, making the installation non-reproducible and exposing users to supply-chain risk from any future malicious or vulnerable release. Because `npx` executes package code directly, a compromised publish could immediately affect users following the README.

External Model or Provider Selection

Medium
Category
Excessive Agency
Content
---
name: deploy-agent-token
description: Deploy an agent token with a Uniswap V4 pool — handles pool creation with configurable hooks (anti-snipe, dynamic fees, revenue share), initial liquidity bootstrapping, LP locking, and post-deployment monitoring. Use when the user wants to launch a token on Uniswap.
model: opus
allowed-tools: [Task(subagent_type:token-deployer)]
---
Confidence
90% confidence
Finding
Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.

Static analysis

No suspicious patterns detected.