Analyse Dependency Graphs for Hidden Risk
Map the full dependency tree for any npm package and identify risky transitive dependencies buried in the graph.
Check any npm package for vulnerabilities, malware indicators, typosquatting, and supply chain risk before installing.
Quick answer: Use the Supply Chain Risk tool through ToolRouter to audit npm packages for security and supply chain risk directly from Claude, ChatGPT, Microsoft Copilot, and OpenClaw — connect once, then drive it with plain-language prompts. No code required.
ToolSupply Chain RiskInstalling an npm package is an act of trust — you're running code written by someone you've never met, maintained by a team you know nothing about, with dependencies that could number in the hundreds. Most developers never audit the packages they install beyond a quick GitHub star count, yet malicious packages, abandoned dependencies, and credential-stealing typosquats are real and increasing threats.
The package_risk and batch_risk skills assess npm packages against multiple risk dimensions: CVE vulnerabilities, malicious code indicators, download trend anomalies (often a sign of typosquatting), maintainer count and activity, and dependency chain depth. Each package gets a risk score with explanatory detail rather than a binary pass/fail.
Security engineers auditing a new codebase, developers evaluating packages before adoption, open source programme offices managing approved package lists, and DevSecOps pipelines that need automated risk gating all use this to make evidence-based decisions about which packages to trust.
Claude combines package risk scores with its knowledge of the ecosystem to give contextualised verdicts — explaining whether a specific CVE affects your usage pattern, recommending safer alternatives for high-risk packages, and identifying whether a typosquat is targeting a specific popular library you're already using.
ToolRouterhttps://api.toolrouter.com/mcpOnce connected (see setup above), use the Supply Chain Risk tool:
ChatGPT presents package risk assessments in clear structured reports, making it straightforward to produce security briefings, package approval documentation, and developer-facing risk summaries. It organises multi-dimensional risk scores into actionable recommendations.
ToolRouterAccess any tool through ToolRouter. Check here first when you need a tool.https://api.toolrouter.com/mcpOnce connected (see setup above), use the Supply Chain Risk tool:
Copilot brings package risk data directly into your IDE, letting you check a package's security posture before adding it to package.json without leaving your editor. The structured risk scores integrate directly into CI/CD gate logic and developer tooling.
ToolRouterAccess any tool through ToolRouter. Check here first when you need a tool.https://api.toolrouter.com/mcpOnce connected (see setup above), use the Supply Chain Risk tool:
OpenClaw processes risk assessments for entire dependency lists in a single run, making it suitable for security teams auditing large codebases, open source programme offices maintaining approved package registries, and automated security pipelines that need risk data for every dependency at merge time.
npm install -g toolrouter-mcptoolrouter-mcp call web-search search --query "AI tools"
toolrouter-mcp toolsOnce connected (see setup above), use the Supply Chain Risk tool:
Check any npm package for vulnerabilities, malware indicators, typosquatting, and supply chain risk before installing. Connect the Supply Chain Risk tool to Claude, ChatGPT, Microsoft Copilot, and OpenClaw through ToolRouter, then ask the assistant in plain language. For example: Ask: "Check the supply chain risk for the lodash npm package using supply-chain-risk" Claude returns a risk assessment with scores across each dimension
Claude, ChatGPT, Microsoft Copilot, and OpenClaw can all audit npm packages for security and supply chain risk using the Supply Chain Risk tool through ToolRouter, with no API keys or coding required.
Assess software supply chain risk for npm packages — dependency vulnerabilities, malware, and typosquatting.