Production-ready AI skills that transform N|Solid MCP data into automated investigation workflows used by NodeSource consultants.
Install any skill into your project using the Vercel Skills CLI:
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-vulnerabilities
npx skills@latest add NodeSource/nsolid-ai-skills/ns-generate-sbom
npx skills@latest add NodeSource/nsolid-ai-skills/ns-audit-dependencies
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-event
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-asset
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-cpu
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-memory
npx skills@latest add NodeSource/nsolid-ai-skills/ns-advanced-memory-leak-hunter
npx skills@latest add NodeSource/nsolid-ai-skills/ns-benchmark-run
npx skills@latest add NodeSource/nsolid-ai-skills/ns-benchmark-validate
npx skills@latest add NodeSource/nsolid-ai-skills/ns-analyze-tracing
npx skills@latest add NodeSource/nsolid-ai-skills/ns-node-upgrade
npx skills@latest add NodeSource/nsolid-ai-skills/ns-upgrade-package
npx skills@latest add NodeSource/nsolid-ai-skills/ns-replace-packageOr manually copy the <skill-name>/SKILL.md file into your project's .claude/skills/ directory.
| Skill | Description |
|---|---|
| ns-analyze-vulnerabilities | Scans running production memory for actively-exploitable CVEs using live N |
| ns-generate-sbom | Generates SPDX/JSON Software Bill of Materials from live running processes |
| ns-audit-dependencies | Audits direct and transitive npm dependencies and produces an NCM-grounded remediation plan |
| Skill | Description |
|---|---|
| ns-analyze-event | Investigates an existing N |
| ns-analyze-asset | Analyzes an existing asset from an asset ID or local file path without capturing a new profile |
| Skill | Description |
|---|---|
| ns-analyze-cpu | Captures V8 CPU profiles, extracts live source code, and identifies bottleneck functions |
| ns-analyze-memory | Diagnoses memory issues via real-time heap sampling and snapshot analysis |
| ns-advanced-memory-leak-hunter | Multi-phase baseline-vs-peak delta analysis for elusive memory leaks |
| ns-benchmark-run | Benchmarks a single Node.js function to measure throughput (ops/sec) using live V8 source or user-provided code |
| ns-benchmark-validate | Scientifically controlled A/B benchmarks with statistical validation (p-value, ops/sec) |
| Skill | Description |
|---|---|
| ns-analyze-tracing | Maps distributed OpenTelemetry spans to diagnose microservice latency and topology issues |
| Skill | Description |
|---|---|
| ns-upgrade-package | Advises on upgrading a specific npm package with risk assessment and rollback guidance |
| ns-replace-package | Compares realistic npm package alternatives and provides migration guidance grounded in NCM data |
| Skill | Description |
|---|---|
| ns-node-upgrade | Recommends the right Node.js upgrade target using authoritative release schedule data and project version detection |
Helper scripts used internally by skills to bridge MCP data and the local filesystem. These helpers are bundled inside the skill directories that reference them.
Downloads a full N|Solid asset (CPU profile, heap snapshot, or heap sampling) from the console API and saves it to .nsolid/assets/. Skills invoke this after triggering a profile/snapshot so the file is available for local tooling (e.g. Chrome DevTools, VS Code memory profiler).
node "<skill-dir>/fetch-asset.cjs" <assetId> <assetType> [appName]| Argument | Description |
|---|---|
assetId |
Asset ID returned by the MCP profile/snapshot/heap-sampling tool |
assetType |
One of: cpuprofile, heapprofile, heapsnapshot |
appName |
(Optional) Application name used in the output path, defaults to unknown |
Output: .nsolid/assets/<assetType>-<appName>-<assetIdPrefix>.<ext>
Reads nsolid.apiBaseUrl and nsolid.authToken from .vscode/settings.json in the workspace root.
Persists a markdown report to .nsolid/assets/ and updates
.nsolid/assets/reports-index.json so the report can be discovered by host
extensions or other local tooling.
node "<skill-dir>/save-report.cjs" <type> <title> <markdown-file> [appName]Output: .nsolid/assets/<type>-<timestamp>.md
Each skill is a standalone, self-contained SKILL.md file compatible with:
- Claude Code (CLI, Desktop, VS Code, JetBrains)
- Windsurf (Cascade)
- Cursor
- GitHub Copilot
- Gemini CLI
- Any tool supporting the Vercel Agent Skills standard
All skills require the N|Solid Console MCP server to be configured and connected to your AI assistant.
Each skill is independent and self-contained — no delegation chains or sub-skill loading required. The AI assistant acts as the natural orchestrator, chaining skills together based on context. For example, after ns-analyze-cpu identifies a bottleneck and proposes a fix, it directs the assistant to invoke ns-benchmark-validate to prove the improvement.