Security audit

Indigo Loan

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Indigo Protocol CDP helper that documents financially sensitive actions, but it does not include hidden code, persistence, or automatic transaction signing.

Install only if you intend to use Indigo Protocol CDP workflows and trust the configured Indigo MCP server. Review every generated transaction in your wallet before signing, especially withdrawals, minting, leverage, liquidation, redemption, freezing, merging, and closing, because signed on-chain transactions can affect funds and CDP safety.

A.I.G

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

SkillSpector

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The skill advertises powerful financial actions such as opening, leveraging, liquidating, freezing, merging, minting, and withdrawing from CDPs without any prominent warning that these actions can cause irreversible loss of funds, liquidation, or other material financial harm. In an agent context, omission of safety guidance increases the chance that users or downstream agents invoke destructive operations without understanding the risks or confirming intent.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: This reference documents numerous high-impact financial write operations such as opening, leveraging, minting, withdrawing, liquidating, redeeming, freezing, and closing CDPs, but it provides no user-facing risk warnings, confirmation guidance, or cautions about irreversible loss, liquidation, or debt exposure. In an agent-tooling context, this omission can cause users or downstream agents to invoke dangerous operations without understanding that even unsigned transactions can trigger significant financial harm once signed.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill documents how to withdraw collateral and close CDPs but does not prominently warn users about liquidation risk, health-ratio degradation, or that financial losses may be irreversible once signed and submitted on-chain. In a DeFi borrowing context, insufficient warnings can lead users to approve transactions that make their position unsafe or permanently realize losses.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: This skill enables users to open leveraged CDP positions, which directly exposes their funds to liquidation and amplified losses, yet the primary description does not prominently warn about those consequences before the action is presented. In a financial skill that can prepare transactions affecting real assets, missing up-front risk disclosure increases the chance that users initiate dangerous actions without understanding downside scenarios.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The example prompts encourage opening leveraged positions but do not directly warn that the operation affects user funds and can trigger liquidation or loss. Because examples strongly shape user behavior, omission of a fund-impact warning in this context makes it more likely that users will treat the action as routine rather than a high-risk financial transaction.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 89% confidence
Finding: The skill exposes high-impact financial actions such as liquidation, redemption, freezing, and merging without clear, repeated warnings that these operations can materially alter positions, consume assets, or become difficult or impossible to reverse after on-chain submission. In a DeFi/CDP context, users may trigger loss-bearing actions or freeze/merge the wrong position if the assistant presents them as routine workflows.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The example prompts encourage users to 'find liquidatable CDPs' and 'liquidate the most undercollateralized CDP' without warning that liquidation may target third-party positions and requires careful verification of protocol eligibility, debt coverage, and financial consequences. This is especially dangerous in a liquidation-focused skill because it normalizes acting on externally owned positions and may lead users to execute profitable but harmful actions without understanding the risks or ethical/legal implications.
Content: No source excerpt is available for this finding.

Vague Triggers

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: This markdown file includes example prompts such as "Analyze the health of my CDP" and "Show me all CDPs for my wallet," but it does not state whether these are the only triggers, provide exclusion conditions, or define contextual limits. Because the skill is described through natural-language prompts alone, invocation boundaries are somewhat ambiguous.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.