ClawHub
Back to skill

Security audit

Silverback Defi

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed DeFi API skill that sends user-directed requests to Silverback and may trigger small USDC payments, with no hidden persistence or destructive behavior in the artifact.

Install only if you are comfortable sending DeFi queries and any supplied wallet addresses to Silverback's service and paying per x402 call in USDC. Prefer using free pricing/endpoint checks first, set wallet spend limits, and avoid the optional global MCP install unless you have reviewed or pinned the package version.

A.I.G

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
SKILL.md:149
Finding

Unpinned Global Installation of a Third-Party npm Package

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:149-153
Vulnerability Type: Insecure third-party dependency installation
Risk Level: Medium

Vulnerable Code

markdown
## MCP Server

For Claude Desktop, Cursor, or Claude Code:
```bash
npm install -g silverback-x402-mcp
text

### Technical Analysis

The documented command installs the latest available version of `silverback-x402-mcp` globally without an exact version pin, lockfile, checksum, signature, or other integrity verification. The package implementation is not included in the audited project, so its behavior cannot be reviewed from the available artifact.

npm packages may execute lifecycle scripts during installation. Because the command uses `-g`, package files and executable entry points are installed outside the project boundary. The command therefore makes the effective installed payload dependent on mutable external registry content and increases the consequences of package compromise, malicious release publication, or unexpected upstream changes.

This finding does not establish that the named package is currently malicious. It identifies an unsafe dependency-installation pattern and the resulting supply-chain exposure.

### Attack Path

1. An attacker compromises the npm publisher account, registry delivery path, or a future release of `silverback-x402-mcp`.
2. The attacker publishes a malicious package version or adds malicious lifecycle or runtime behavior.
3. A user follows the Skill's unpinned `npm install -g` instruction.
4. npm resolves and downloads the mutable latest package release.
5. Malicious lifecycle scripts may execute during installation, or the installed MCP server may execute attacker-controlled logic when subsequently launched by an AI client.
6. The payload operates with the permissions and data access available to the installing user or MCP process.

### Impact Assessment

Successful exploitation could pe
...[truncated 576 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin the dependency to an exact, reviewed version rather than resolving the latest release:
    bash
    npm install --save-exact silverback-x402-mcp@<reviewed-version>
  2. Prefer a project-local installation over -g to limit filesystem scope and make dependency state reproducible.
  3. Commit a lockfile and verify registry integrity metadata in controlled deployments.
  4. Review the package source and all transitive dependencies before recommending installation.
  5. Disable lifecycle scripts where the package can function without them:
    bash
    npm install --save-exact silverback-x402-mcp@<reviewed-version> --ignore-scripts
  6. If lifecycle scripts are required, document them explicitly and execute installation in a sandbox or least-privileged environment.
  7. Publish and verify package provenance, signatures, or checksums, and establish a controlled update process for future versions.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

This skill instructs the agent to send user-provided prompts and query data to an external third-party service over the network. That creates a real data exfiltration surface: chat prompts, token interests, wallet addresses, and trading intent may be transmitted off-platform, which is especially sensitive in a DeFi context where such information can reveal financial behavior.

Content

Scanner excerpt · SKILL.md (reported line 22)May include surrounding context.

AI chat with all 19 intelligence tools. Ask any DeFi question.

bash
curl -s -X POST https://x402.silverbackdefi.app/api/v1/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "What are the top coins right now?"}'

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly states that paid endpoints use x402 and USDC, but it does not present a clear, prominent safety warning that invoking these endpoints can spend real funds from the agent's configured wallet. In an agent setting, users may treat examples as harmless API calls, so insufficient disclosure can lead to unintended micropayments and cumulative financial loss.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.