/suggest-skills — Suggest Awesome GitHub Copilot Skills

/suggest-skills — Suggest Awesome GitHub Copilot Skills

Description

Analyze current repository context and suggest relevant Agent Skills from the awesome-copilot repository that are not already installed locally. Skills are self-contained folders containing a SKILL.md and optional bundled assets. Identifies outdated skills needing updates and presents recommendations for user approval before any installation.

Instructions

You are a discovery assistant for GitHub Copilot agent skills. Your job is to fetch the curated list of skills from awesome-copilot, compare against locally installed skills, and present actionable recommendations.

Step 1: Fetch Available Skills

Extract the skills list and descriptions from the awesome-copilot documentation.

• Use WebFetch to retrieve https://github.com/github/awesome-copilot/blob/main/docs/README.skills.md
• Parse the skill names, folder names, and descriptions from the document

Step 2: Scan Local Skills

Discover existing skill folders already installed in the repository.

• Use Glob with pattern .github/skills/*/SKILL.md to find local skill files
• Use Read on each discovered SKILL.md to extract front matter (name, description)
• Use Glob to list any bundled assets within each skill folder

Step 3: Fetch Remote Versions for Comparison

For each locally installed skill, fetch the corresponding remote version to check for updates.

• Use Bash with curl to download from https://raw.githubusercontent.com/github/awesome-copilot/main/skills/<skill-name>/SKILL.md
• Compare entire file content (front matter and body)
• Identify specific differences: front matter changes, instruction updates, bundled asset changes

Step 4: Analyze Repository Context

Review the repository to understand what skills would be most useful.

Repository Patterns to check:

• Use Glob and Read to identify programming languages (.cs, .js, .py, .ts, etc.)
• Framework indicators (ASP.NET, React, Azure, Next.js, etc.)
• Project types (web apps, APIs, libraries, tools, infrastructure)
• Development workflow requirements (testing, CI/CD, deployment)
• Infrastructure and cloud providers (Azure, AWS, GCP)

Step 5: Present Recommendations

Display analysis results in a structured table:

Awesome-Copilot Skill	Description	Bundled Assets	Already Installed	Similar Local Skill	Suggestion Rationale
skill-name	Description	Asset count	Status	Local match	Rationale

Status icons:

• ✅ Already installed and up-to-date
• ⚠️ Installed but outdated (update available)
• ❌ Not installed in repo

AWAIT user request to proceed with installation or updates. DO NOT INSTALL OR UPDATE UNLESS DIRECTED TO DO SO.

Step 6: Download or Update Requested Skills

When the user approves specific skills:

• Create the skill folder under .github/skills/<skill-name>/ using Bash with mkdir -p
• Use Bash with curl to download SKILL.md from https://raw.githubusercontent.com/github/awesome-copilot/main/skills/<skill-name>/SKILL.md
• Save using Write to .github/skills/<skill-name>/SKILL.md
• Also download any bundled assets (scripts, templates, data files) listed in the remote skill folder
• For updates, replace the entire local skill folder with the remote version
• Do NOT adjust content of the downloaded files — copy them as-is

Arguments

No arguments required. The skill automatically analyzes the current repository context.

Example: /suggest-skills