### Add Memory - Setup Knowledge

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/add_memory/instructions.md

Use this snippet to document critical project setup, installation, or configuration steps. This is particularly useful for remembering essential commands or procedures that are easy to forget.

```javascript
add_memory({
  content: "Run 'npm run db:push' to sync Prisma schema changes to database. Required after schema modifications before server restart.",
  category: "SETUP",
  explanation: "Critical setup step that's easy to forget"
})
```

--------------------------------

### Install Dependencies

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Clone the repository and install project dependencies using npm.

```bash
git clone <repository-url>
cd shadow
npm install
```

--------------------------------

### Start Development Servers

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Start all development services or specific services using npm run dev with filters.

```bash
# Start all services
npm run dev

# Or start specific services
npm run dev --filter=frontend
npm run dev --filter=server
npm run dev --filter=sidecar
```

--------------------------------

### List Memories - Architecture Understanding

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/list_memories/instructions.md

Employ this example to access past architectural patterns and decisions before introducing new features. A specific explanation helps in retrieving relevant information.

```javascript
list_memories({
  explanation: "Understanding existing patterns before adding new API endpoints"
})
```

--------------------------------

### Manage Shadow Infrastructure Components

Source: https://github.com/ishaan1013/shadow/blob/main/scripts/README.md

Scripts for deploying specific infrastructure components or cleaning up the entire setup. Use these for targeted deployments or removal of resources.

```bash
# Deploy only EKS cluster with Kata Containers
./scripts/deploy-remote-infrastructure.sh
```

```bash
# Deploy only ECS backend service
./scripts/deploy-backend-ecs.sh
```

```bash
# Deploy complete infrastructure (EKS + ECS)
./scripts/deploy-full-infrastructure.sh
```

```bash
# Clean up infrastructure
./scripts/cleanup-infrastructure.sh
```

--------------------------------

### Non-interactive npm Install

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/run_terminal_cmd/instructions.md

Use the --yes flag for non-interactive package installations to avoid prompts.

```bash
npm install --yes
```

--------------------------------

### List Memories - Starting New Task

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/list_memories/instructions.md

Use this snippet to retrieve memories when beginning a new task, providing context for the upcoming work. Include a clear explanation of why memories are needed.

```javascript
list_memories({
  explanation: "Reviewing repository context before implementing user authentication system"
})
```

--------------------------------

### Update Todo List Status

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/todo_write/instructions.md

Use this to update the status of existing tasks, such as marking a task as completed and starting a new one. Set merge to true to add to or modify the existing list.

```javascript
// When starting a new task
todo_write({
  merge: true,
  todos: [
    {id: "auth", content: "Implement user authentication", status: "completed"},
    {id: "api", content: "Create REST API endpoints", status: "in_progress"}
  ],
  explanation: "Updating status as I begin API implementation"
})
```

--------------------------------

### Regex Escaping Examples

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/grep_search/instructions.md

Special characters in regex patterns must be escaped to be treated as literal characters. This ensures accurate matching of symbols, function calls, file names, and logical operators.

```regex
function\(
```

```regex
value\[index\]
```

```regex
file\.txt
```

```regex
user\|admin
```

--------------------------------

### Deploy Shadow Application

Source: https://github.com/ishaan1013/shadow/blob/main/scripts/README.md

Execute the script to deploy the complete Shadow platform after the infrastructure is set up.

```bash
# Deploy complete platform
./scripts/deploy-full-infrastructure.sh
```

--------------------------------

### Build Applications

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Build all project packages and applications, or target specific applications for building.

```bash
# Build all packages and apps
npm run build
```

```bash
# Build specific app
npm run build --filter=frontend
```

```bash
npm run build --filter=server
```

```bash
npm run build --filter=sidecar
```

--------------------------------

### Create Initial Todo List

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/todo_write/instructions.md

Use this to create a new, comprehensive task list for a complex project. Set merge to false to replace any existing todos.

```javascript
todo_write({
  merge: false,
  todos: [
    {id: "setup", content: "Set up project structure", status: "completed"},
    {id: "auth", content: "Implement user authentication", status: "in_progress"},
    {id: "api", content: "Create REST API endpoints", status: "pending"},
    {id: "frontend", content: "Build React frontend", status: "pending"},
    {id: "tests", content: "Write unit tests", status: "pending"}
  ],
  explanation: "Creating comprehensive task list for user management system"
})
```

--------------------------------

### Configure Local PostgreSQL Database

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Create a local PostgreSQL database named 'shadow_dev' and update the database URL in packages/db/.env. Then, generate the Prisma client and push the schema.

```bash
# Create local PostgreSQL database
psql -U postgres -c "CREATE DATABASE shadow_dev;"

# Update packages/db/.env with your database URL
DATABASE_URL="postgres://postgres:@127.0.0.1:5432/shadow_dev"

# Generate Prisma client and push schema
npm run generate
npm run db:push
```

--------------------------------

### Frontend Environment Configuration

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Configure environment variables for the frontend, including the server URL, Vercel environment, and GitHub authentication details.

```bash
# Point frontend to your server if needed
NEXT_PUBLIC_SERVER_URL=http://localhost:4000

# Marks environment; any value other than "production" enables local behavior
NEXT_PUBLIC_VERCEL_ENV=development

GITHUB_PERSONAL_ACCESS_TOKEN=ghp_xxx
GITHUB_CLIENT_ID=
GITHUB_CLIENT_SECRET=
```

--------------------------------

### Configure AWS SSO and Deploy Remote Infrastructure

Source: https://github.com/ishaan1013/shadow/blob/main/scripts/README.md

Configure AWS Single Sign-On for the 'ID' profile and then deploy the remote infrastructure, which includes the EKS cluster with Kata Containers. This process can take 25-35 minutes.

```bash
# Configure AWS SSO
aws configure sso --profile=ID

# Deploy infrastructure (25-35 minutes)
./scripts/deploy-remote-infrastructure.sh
```

--------------------------------

### Database Troubleshooting

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Commands to reset the development database and regenerate the Prisma client.

```bash
# Reset development database
npm run db:push:reset

# Regenerate Prisma client
npm run generate
```

--------------------------------

### Server Environment Configuration

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Configure required and optional environment variables for the server, including database URL, authentication secret, GitHub token, execution mode, and Pinecone settings.

```bash
# Required
DATABASE_URL="postgres://postgres:@127.0.0.1:5432/shadow_dev"
BETTER_AUTH_SECRET="dev-secret"

GITHUB_PERSONAL_ACCESS_TOKEN=ghp_xxx

# Local mode
NODE_ENV=development
AGENT_MODE=local

# Optional: Pinecone for semantic search
PINECONE_API_KEY="" # TODO: Set this to your Pinecone API key
PINECONE_INDEX_NAME="shadow"

# Workspace directory for local agent:
WORKSPACE_DIR= # TODO: Set this to your local workspace directory
```

--------------------------------

### Copy Environment Templates

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Copy template environment files to create new environment configuration files for server, frontend, and database.

```bash
cp apps/server/.env.template apps/server/.env
cp apps/frontend/.env.template apps/frontend/.env
cp packages/db/.env.template packages/db/.env
```

--------------------------------

### Run Project Checks

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Execute type checking and linting for the project.

```bash
npm run check-types
npm run lint
```

--------------------------------

### Run Linting and Formatting Checks

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Execute linting, formatting, and type checking across all packages and applications within the project.

```bash
# Lint all packages and apps
npm run lint
```

```bash
# Format code with Prettier
npm run format
```

```bash
# Type checking
npm run check-types
```

--------------------------------

### Deploy Infrastructure Scripts

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Scripts for deploying and cleaning up cloud infrastructure, including EKS clusters and ECS services.

```bash
# Deploy EKS cluster with Kata Containers
./scripts/deploy-remote-infrastructure.sh

# Deploy ECS backend service  
./scripts/deploy-backend-ecs.sh

# Infrastructure cleanup
./scripts/cleanup-infrastructure.sh
```

--------------------------------

### Set Development Environment Variables

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Configure local development environment variables, including authentication secrets and database connection strings.

```bash
BETTER_AUTH_SECRET=dev-secret
```

```bash
DATABASE_URL="postgres://postgres:@127.0.0.1:5432/shadow_dev"
DIRECT_URL="postgres://postgres:@127.0.0.1:5432/shadow_dev"
```

--------------------------------

### Manage Database Operations with Prisma

Source: https://github.com/ishaan1013/shadow/blob/main/README.md

Commands for database schema generation, migration, and management using Prisma.

```bash
# Generate Prisma client from schema
npm run generate
```

```bash
# Push schema changes to database (for development)
npm run db:push
```

```bash
# Reset database and push schema (destructive)
npm run db:push:reset
```

```bash
# Open Prisma Studio GUI
npm run db:studio
```

```bash
# Run migrations in development
npm run db:migrate:dev
```

--------------------------------

### Build and Type Checking

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Commands to resolve module resolution issues by rebuilding the project and to check for type errors.

```bash
# Module resolution issues
npm run build

# Type checking issues
npm run check-types
```

--------------------------------

### Server Testing Commands

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Commands for running type checking and linting on the server project using npm filters.

```bash
# Server testing  
npm run check-types --filter=server
npm run lint --filter=server
```

--------------------------------

### Database Schema Management

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Commands for pushing development database schema changes and applying production migrations.

```bash
# Development
npm run db:push

# Production migrations
npm run db:migrate:dev
npm run db:migrate:deploy
```

--------------------------------

### Frontend Testing Commands

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Commands for running type checking and linting on the frontend project using npm filters.

```bash
# Frontend testing
npm run check-types --filter=frontend
npm run lint --filter=frontend
```

--------------------------------

### Add Memory - Architecture Pattern

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/add_memory/instructions.md

Use this snippet to record key architectural decisions, system design, component relationships, or data flow. This helps in remembering critical design choices for future development.

```javascript
add_memory({
  content: "Uses Turborepo monorepo with shared packages. Frontend (Next.js) communicates with server via WebSocket for real-time updates. Agent execution can run in local or remote mode via tool executor abstraction.",
  category: "ARCHITECTURE", 
  explanation: "Recording key architectural decisions for future development"
})
```

--------------------------------

### Database Model Relationships in TypeScript

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Illustrates the one-to-many and many-to-one relationships between core database models like User, Task, ChatMessage, and Memory.

```typescript
User -> Task (one-to-many)
Task -> ChatMessage (one-to-many) 
Task -> Memory (one-to-many)
Task -> Todo (one-to-many)
Task -> CodebaseUnderstanding (many-to-one)
```

--------------------------------

### Add Memory - Bug Pattern

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/add_memory/instructions.md

Use this snippet to document known issues, gotchas, or troubleshooting steps related to specific components or behaviors. This helps in avoiding future confusion and speeds up debugging.

```javascript
add_memory({
  content: "Checkbox onCheckedChange receives boolean | 'indeterminate' but can cast to boolean for simple toggles. Radix UI never actually passes indeterminate for basic checkboxes.",
  category: "BUGS",
  explanation: "Documenting UI component behavior to avoid future confusion"
})
```

--------------------------------

### Find Import Statements

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/grep_search/instructions.md

Use this pattern to find import statements, particularly those importing from React. It matches 'import', followed by any characters, 'from', any characters, and 'react'.

```regex
import.*from.*react
```

--------------------------------

### List Memories - Debugging Context

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/list_memories/instructions.md

Utilize this snippet for debugging purposes to check for known issues or solutions from previous sessions. Provide a concise explanation related to the problem being investigated.

```javascript
list_memories({
  explanation: "Checking for known issues related to database connection problems"
})
```

--------------------------------

### Find Function Definitions

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/grep_search/instructions.md

This regex pattern helps identify function definitions in the code. It specifically looks for the keyword 'function' followed by a space and a function name.

```regex
function myFunction\(
```

--------------------------------

### Find TODO Comments

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/grep_search/instructions.md

Use this pattern to locate all lines containing 'TODO' comments within the codebase. This is useful for tracking outstanding tasks or issues.

```regex
TODO:
```

--------------------------------

### Remove Memory After Architecture Change

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/remove_memory/instructions.md

Use this snippet to remove memories related to previous architectural patterns that are no longer applicable. Provide a clear explanation for the removal.

```javascript
remove_memory({
  memoryId: "mem_def456", 
  explanation: "Removed old WebSocket implementation, this pattern no longer applies"
})
```

--------------------------------

### Avoid Pager Output

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/run_terminal_cmd/instructions.md

Append | cat to commands that might produce pager output to ensure all content is displayed.

```bash
npm test | cat
```

--------------------------------

### Run Long-Running Commands in Background

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/run_terminal_cmd/instructions.md

Set is_background=true for commands that are expected to run for an extended period.

```bash
npm run dev with is_background=true
```

--------------------------------

### Remove Incorrect Memory

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/remove_memory/instructions.md

Use this snippet to remove memories containing misinformation or misidentified patterns. The `explanation` should clarify the correction.

```javascript
remove_memory({
  memoryId: "mem_ghi789",
  explanation: "This bug pattern was misidentified, actual issue was different"
})
```

--------------------------------

### Remove Outdated Memory

Source: https://github.com/ishaan1013/shadow/blob/main/apps/server/src/agent/tools/prompts/remove_memory/instructions.md

Use this snippet to remove a memory that is no longer accurate due to code changes. Ensure the `memoryId` is correct.

```javascript
remove_memory({
  memoryId: "mem_abc123",
  explanation: "Database schema has changed, this setup information is outdated"
})
```

--------------------------------

### Message Part Types in TypeScript

Source: https://github.com/ishaan1013/shadow/blob/main/CLAUDE.md

Defines the discriminated union types for structured message parts used in real-time streaming via Socket.IO.

```typescript
// Message part types in streaming
- TextPart: Regular text content
- ReasoningPart: AI reasoning with signatures
- ToolCallPart: Tool execution requests
- ToolResultPart: Tool execution results
- ErrorPart: Error information
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.