Skip to main content
Version: 6.5.0

Agent Skills

N|Solid in the IDE can use agent skills from the NodeSource AI Toolkit to make its AI workflows more structured, consistent, and useful. These skills give the AI specific guidance for common N|Solid tasks such as investigating runtime events, analyzing assets, benchmarking functions, and reviewing dependency risk.

What Skills Are​

In N|Solid in the IDE, a skill is a packaged set of instructions and workflow context that helps the AI handle a specific job. Rather than asking a model to improvise from scratch every time, the extension can load the relevant skill and use it to guide the analysis.

This is especially helpful for tasks that need a repeatable structure, such as:

  • Analyzing a CPU issue
  • Investigating a runtime event
  • Reviewing a downloaded diagnostic asset
  • Auditing dependencies
  • Validating an optimization with benchmarking

Installing The AI Toolkit Skills​

Skills are installed as part of the NodeSource AI Toolkit setup flow in N|Solid in the IDE.

From the Welcome experience, users can install or reinstall the toolkit. During that process, N|Solid in the IDE:

  • Creates or updates the workspace MCP configuration in .vscode/mcp.json
  • Installs or updates the skills in .agents/skills

This setup prepares the workspace so that N|Solid in the IDE can use both the MCP-backed tools and the skill definitions that guide AI behavior.

Where Skills Live​

After installation, the skills are stored in the workspace under:

.agents/skills

This makes the toolkit local to the workspace and allows N|Solid in the IDE to load the relevant skill content when a user starts one of the supported AI workflows.

What Skills Power In N|Solid in the IDE​

The AI Toolkit skills support several of the main extension features:

  • Asset analysis -- Skills help the AI inspect downloaded N|Solid assets such as profiles and heap snapshots.
  • Event investigation -- Skills help analyze runtime event payloads and recommend next steps.
  • CPU analysis -- Skills guide investigation of CPU-related performance issues.
  • Memory analysis -- Skills support heap and memory-oriented diagnostics.
  • Benchmarking -- Skills drive function benchmarking and optimization validation workflows.
  • Dependency review -- Skills support package audits, package upgrades, and replacement recommendations.
  • Node.js upgrade guidance -- Skills help structure runtime upgrade advice.

Agent Skills

Common Skill-Driven Workflows​

Some of the key skill-backed workflows available in N|Solid in the IDE include:

  • analyze-asset -- Analyze an existing N|Solid asset or local profile file.
  • analyze-event -- Investigate an N|Solid event with runtime context.
  • analyze-cpu -- Investigate CPU-related performance problems.
  • analyze-memory -- Investigate memory behavior and related runtime issues.
  • benchmark-run -- Benchmark a single function and save the result.
  • benchmark-validate -- Compare original and optimized implementations with A/B benchmarks.
  • audit-dependencies -- Support dependency audit and remediation workflows.
  • upgrade-package -- Help evaluate and guide a package upgrade.
  • replace-package -- Help identify a safer or more maintainable replacement package.
  • node-upgrade -- Support Node.js runtime upgrade planning.

These workflows are usually triggered through the N|Solid chat experience or from extension views that hand context into chat for the user.

NodeSource Support