Skip to content

■ SKILLS // SKILL SPECS

auggie-context

Context management and enhancement plugin for Claude Code

■ [!] ON THIS PAGE ▼
■ INSTALL // npx @self.md/cli install auggie-context
Author1WP
Namespace@vleytman/1wp-plugins
Category#productivity
Version0.1.0
Stars★ 1
Downloads↓ 3
Verification[ ✓ ] self.md verified

Context management and enhancement plugin for Claude Code

Installation

npx claude-plugins install @vleytman/1wp-plugins/auggie-context

Contents

Folders: agents, commands, hooks, skills

Files: README.md

Documentation

A context management and codebase exploration plugin for Claude Code, powered by the Auggie SDK and Context Engine SDK .

Features

  • /auggie: Context-aware codebase assistant
  • /auggie-setup: Configure Auggie credentials
  • /auggie-index: Index your codebase for fast context search
  • Auggie Agent: Specialized subagent for codebase exploration
  • Auggie Skill: Automatic context-aware assistance

Getting Started

Step 1: Install the Plugin

# Add the 1wp-plugins marketplace
/plugin marketplace add vleytman/1wp-plugins

# Install the plugin
/plugin install auggie-context@1wp-plugins

Step 2: Set Up Credentials

/auggie-setup

Or manually:

# Install Auggie CLI (if not installed)
# See: https://docs.augmentcode.com/cli

# Login to Auggie
auggie login

# Verify it worked
auggie token print

Restart Claude Code after logging in.

Step 3: Index Your Codebase

/auggie-index

This creates auggie-context.json which persists your index between sessions.

Step 4: Start Using

/auggie How does the authentication system work?
/auggie Find all API endpoints
/auggie Explain the data flow in this component

Commands

CommandDescription
/auggie [query]Ask questions about your codebase
/auggie-setupConfigure Auggie credentials
/auggie-indexIndex or re-index the codebase

How It Works

  1. Authentication: Uses auggie login which stores credentials in ~/.augment/session.json
  2. Indexing: Uses DirectContext API to index files and persist state to auggie-context.json
  3. Search: Queries are processed by Auggie’s context engine for relevant code snippets
  4. Persistence: Index state is saved locally, enabling fast incremental updates

FAQ / Troubleshooting

“Auggie not configured” error

Run /auggie-setup or:

auggie login

Then restart Claude Code.

“Command not found: auggie”

Install the Auggie CLI:

Index seems stale or incomplete

Re-run the indexing command:

/auggie-index

“Not authenticated” or token errors

Your session may have expired. Re-authenticate:

auggie login

How do I check if I’m authenticated?

auggie token print

If it shows accessToken and tenantURL, you’re good.

Where are credentials stored?

Credentials are stored in ~/.augment/session.json (created by auggie login). This file is outside your project and not committed to git.

Where is the index stored?

The index is stored in auggie-context.json in your project root. This file:

  • Is project-specific
  • Should be in .gitignore (added automatically)
  • Is regenerated by /auggie-index

How do I keep the index up to date?

Run /auggie-index after making significant changes to your codebase. The indexing is incremental - only changed files are re-processed.

Large codebase taking too long to index?

  • Ensure you have a .gitignore file to exclude node_modules/, build artifacts, etc.
  • You can also create an .augmentignore file for additional exclusions

License

MIT

Included Skills

This plugin includes 1 skill definition:

auggie

View skill definition

When to use this skill

Whenever you need to explore a codebase, understand architecture, or find specific code patterns. Use this skill when:

  • The user asks questions about the codebase structure
  • You need to find examples of how something is implemented
  • You need to understand the architecture of a project
  • You need to trace code flows or dependencies

How to use this skill

Auggie is a context-aware assistant for codebase exploration.

Capabilities

  1. Code Search: Find files, functions, classes, and patterns
  2. Architecture Analysis: Understand project structure and organization
  3. Example Finding: Locate relevant code examples
  4. Flow Tracing: Follow code paths and dependencies

Best Practices

  • Ask specific questions for better results
  • Provide context about what you’re trying to accomplish
  • Use natural language to describe what you’re looking for

Examples

"How is authentication handled in this project?"
"Find all API endpoints"
"What patterns are used for error handling?"
"Show me how the database models are structured"

Keywords

codebase, context, search, explore, architecture, patterns, examples, code, structure

Source

View on GitHub