Back to skill

Security audit

Deploy Contracts

Security checks for vulnerabilities and agentic risk

Overview

This Aptos deployment guide is mostly coherent, but it includes auto-confirm commands for mainnet deployments and upgrades that can submit high-impact blockchain transactions without an interactive confirmation.

Review this skill before installing if you may use it for mainnet. Do not let an agent run the `--assume-yes` mainnet deploy or upgrade commands unless you have independently checked the network, account, object address, code diff, compatibility, gas cost, and transaction impact. The skill is not malicious, but its safest use requires explicit human confirmation for production transactions.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest description says the skill triggers on phrases including 'how to deploy' and 'deployment checklist', which are broad natural-language phrases that can appear in many ordinary conversations. The trigger list does not include exclusion conditions or context limits, so the activation scope is wider than necessary for a deployment-specific skill.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
92% confidence
Finding

The skill recommends --assume-yes for a mainnet deployment flow, which suppresses interactive confirmation for a real-value blockchain transaction. In an agent-driven context, this increases the risk of unintended or insufficiently reviewed deployments, including wrong-network, wrong-address, or unexpected-cost transactions becoming irreversible on mainnet.

Content

Scanner excerpt · SKILL.md (reported line 265)May include surrounding context.

md
# 1. Gas confirmation: Review gas costs
# 2. Object address: Note the object address for future reference

# OR use --assume-yes to auto-confirm (only if you're confident)
aptos move deploy-object \
    --address-name my_addr \
    --profile default \

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 297)May include surrounding context.

Step 8: Document Deployment

Create deployment record:

markdown
# Deployment Record

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The upgrade example includes an auto-confirm path for upgrade-object, which can cause an irreversible contract upgrade to proceed without an explicit human checkpoint. In a smart-contract deployment skill, bypassing prompts is especially risky because upgrades may alter production logic, break compatibility, or incur real financial and operational consequences.

Content

Scanner excerpt · SKILL.md (reported line 336)May include surrounding context.

md
--object-address <object_address_from_initial_deploy> \
    --profile mainnet

# Upgrade with auto-confirm
aptos move upgrade-object \
    --address-name my_addr \
    --object-address <object_address> \

Static analysis

No suspicious patterns detected.