Back to skill

Security audit

github-repo-due-diligence-lite

Security checks for vulnerabilities and agentic risk

Overview

The skill performs its disclosed public GitHub metadata check, with a documented paid upgrade offer and ordinary installation precautions.

Treat the report as a maintenance indicator, not proof that repository code is safe. Review and pin the installer and source before installation. Expect a paid upgrade suggestion and a request to submit your rating to the marketplace.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied chunk is an installation command, not an implementation of the advertised single-call GitHub health assessment. It does not show retrieval or evaluation of any declared repository metrics. This assessment applies only to the supplied command; the behavior of the installed skill is not available for inspection.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The API access and health summary match the description. However, when fields such as forks, open issues, default branch, or language are present, the code also generates reasons to upgrade to another skill. This promotional behavior is not disclosed and is not necessary to implement the stated health check. The code does not invoke the promoted skill or make additional requests.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The installation command invokes npx skills without a version, allowing npm to download and execute a changing package version with the user's privileges. A compromised future release could therefore execute malicious code during installation; although the finding incorrectly labels this an MCP server reference, the underlying dependency-pinning concern is valid. This is a conventional installation example with no evidence of malicious intent, and the skill's unauthenticated, read-only GitHub check does not mitigate installation-time risk.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/dd.js (reported line 55)May include surrounding context.

js
}
  const [owner, repo] = [m[1], m[2].replace(/\.git$/, "")];
  try {
    const r = await get(`https://api.github.com/repos/${owner}/${repo}`);
    if (r.status !== 200) {
      console.error(JSON.stringify({ error: `GitHub ${r.status}` }));
      process.exit(1);

Static analysis

No suspicious patterns detected.