How to Apply Skills to a Session

By Paul Flanders · · Beginner

Learn how to apply skills to a session for consistent, house-style results without repeating instructions. Discover how to manage and activate skills, including adding team-specific ones, and troubleshoot common issues.

Estimated time: 3 minutes
Difficulty: Beginner

Why you'd use this

A skill is a reusable guide that teaches the assistant how to do one kind of work well, for example building n8n automations or writing Moodle plugins, in the Agent Skills layout of a folder with a SKILL.md. When you apply a skill it is included with every request for the session, so you get consistent, house-style results without pasting instructions each time. The extension ships with a library of more than ninety skills, and you can add your team's own.

Before you start

Permissions required:

  • None for built-in skills. Write access to .project-ai/skills/ to add workspace skills.

You'll need:

  • For your own skills: a folder <name>/SKILL.md with YAML front matter (name and description) followed by markdown instructions, or a single <name>.md file.

Steps

  1. Type /skills in the composer, or press / and choose Skills (apply / remove).
  2. A pick list opens with a checkbox per skill and its description. Tick the ones to enable and press Enter. Untick to remove.
  3. A skills bar appears under the header listing the active skills, with a Manage link to reopen the list.
  4. Ask for work in that domain. When a reply draws on a skill, a "Skill applied" chip appears on it.
  5. To add a team skill, create .project-ai/skills/<name>/SKILL.md in the workspace, or point the ellmCode.skillsPath setting at a shared folder (absolute or workspace-relative). Workspace skills override a shared-library skill of the same name, which overrides a built-in.
  6. In Plan Mode, watch for a "Suggested skills for this project" box after the first planning turn. Tick the suggestions and click Enable selected, or Skip.

What you should see

Active skills are remembered per workspace and are listed in the audit log for every request they were sent with, which is how you confirm a skill was in effect for a given change.

Troubleshooting

  • A new skill file is not listed: check the front matter has a name line and the folder sits directly under the skills path. Reopen the picker after saving.
  • Replies got slower with many skills on: each skill adds to every request. Keep two or three active for the task at hand and remove the rest.

Related guides

  • How to Run a Specialist Agent
  • How to Plan and Build a Feature with Plan Mode
  • How to Show the Audit Log and Debug Trace

person people found this useful.

Related

Knowledge base 11 Sep 2026

How to Send Identity and Intent Headers to the Gateway

Learn how to configure identity and intent headers for a shared gateway using eLLM Code. This guide explains how to adjust settings in VS Code, ensuring requests are routed correctly and usage is attributed accurately.

Knowledge base 11 Sep 2026

How to Tune Generation Settings for Your Model

Learn how to fine-tune generation settings for local and self-hosted models to improve performance and accuracy. Adjust parameters like repetition, prompt size, and command timeouts to better match your model's behaviour.

Knowledge base 11 Sep 2026

How to Run the Benchmark Suite

This article provides a detailed guide on running a benchmark suite to test model configurations, including setup requirements, steps for execution, and troubleshooting tips. It's aimed at advanced users looking to evaluate model capabilities before deployment.

Expand 29 Jul 2026

How AI can support pupils with SEND

Explore how AI tools are aiding pupils with special educational needs and disabilities by simplifying text, breaking down tasks, and enhancing accessibility. Learn about the considerations schools should make before implementation.

Expand 29 Jul 2026

AI policy and practice in further education

Further education colleges need distinct AI policies due to their unique mix of vocational courses, diverse age groups, and specific funding and inspection requirements. This guidance highlights the importance of tailored AI approaches in FE settings.