### Local Development Setup and Execution

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/github-action-indexer/README.md

Clone the repository, install dependencies, set environment variables, and run the indexing and search commands locally.

```bash
git clone https://github.com/augmentcode/auggie.git
cd auggie/examples/typescript-sdk/context/github-action-indexer

npm install

export AUGMENT_API_TOKEN="your-token"
export AUGMENT_API_URL="https://your-tenant.api.augmentcode.com/"
export GITHUB_TOKEN="your-github-token"
export GITHUB_REPOSITORY="owner/repo"
export GITHUB_SHA="$(git rev-parse HEAD)"

npm run index

npm run search "authentication functions"
npm run search "error handling"
```

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

### Install GitHub Action Indexer

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Installs dependencies for the GitHub Action Indexer example.

```bash
npm run github-indexer:install
```

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

### Run FileSystem Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Executes the FileSystem Context example for local directory search via MCP protocol. Ensure Auggie CLI is installed and authenticated.

```bash
npm run filesystem-context
```

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

### Running Auggie SDK Examples

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

These bash commands show how to execute the provided Python examples. You can run a specific example, specify a custom workspace, or select a different model.

```bash
# Run a specific example
python basic_usage.py

# Run with custom workspace
python basic_usage.py --workspace /path/to/workspace

# Run with different model
python basic_usage.py --model gpt-4o

```

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

### Authenticate and Run Direct Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/direct_context/README.md

Shows how to authenticate with Auggie and run the direct context example from the command line.

```bash
# Authenticate
auggie login

# Run the example (from the context directory)
cd examples/python-sdk/context
python -m direct_context

# Or run directly
python direct_context/main.py
```

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

### Run Individual Example with Python SDK

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/user_examples/README.md

Execute a specific Python example script from the command line. Ensure you are in the directory containing the script.

```bash
python 01_quick_start.py
```

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

### Install SDK Dependencies

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Installs project dependencies for the TypeScript SDK context examples.

```bash
cd examples/typescript-sdk/context
npm install
```

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

### Run File Search Server Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Starts a REST API for semantic file search with AI summarization. The server runs locally on port 3000.

```bash
python -m file_search_server .
```

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

### Authenticate and Run Direct Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/direct-context/README.md

Use these commands to authenticate your Auggie client and then run the direct context example.

```bash
# Authenticate
auggie login

# Run the example
npx tsx examples/context/direct-context/index.ts
```

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

### Run Direct Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Executes the Direct Context example, which uses API-based indexing with semantic search and AI Q&A.

```bash
python -m direct_context
```

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

### Install Auggie CLI

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Installs the Auggie CLI globally. This is a prerequisite for FileSystem Context examples.

```bash
npm install -g @augmentcode/auggie@prerelease
```

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

### Run Direct Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Executes the Direct Context example, which uses API-based indexing with semantic search and AI Q&A.

```bash
npm run direct-context
```

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

### Prompt Example: Documentation Generator

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

An example prompt for generating documentation for public functions and classes, including validation against the Google style guide.

```plaintext
Analyze all public functions and classes in the src/ directory. For each
one that's missing a docstring or has an incomplete docstring, generate
comprehensive documentation including description, parameters, return
values, and usage examples. Then validate that all docstrings follow
Google style guide format.
```

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

### Run Prompt Enhancer Server Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Starts an HTTP server that enhances prompts with codebase context. The server runs locally on port 3001.

```bash
python -m prompt_enhancer_server .
```

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

### Run Basic Usage Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Execute the basic_usage.py script to demonstrate fundamental SDK features.

```bash
python basic_usage.py
```

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

### Quick Start with Auggie SDK

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/user_examples/user_guide.md

Instantiate the Auggie agent and run a simple instruction.

```python
from auggie_sdk import Auggie

# Create an agent instance
agent = Auggie()

# Run a simple instruction
response = agent.run("What is the capital of France?")
print(response)
# Output: Paris
```

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

### Install and Authenticate Auggie CLI

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/file_search_server/README.md

Install the Auggie CLI globally and authenticate your account to use the service.

```bash
npm install -g @augmentcode/auggie@prerelease
auggie login
```

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

### Run Function Calling Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/FUNCTION_CALLING.md

Navigate to the example directory and execute the Python script to test the function calling capabilities of the Auggie SDK.

```bash
cd examples/python-sdk
python3 function_calling_example.py
```

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

### Authenticate and Run Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/filesystem_context/README.md

Authenticate with Auggie for AI features and then run the filesystem context example from the specified directory. Alternatively, you can run the main Python script directly.

```bash
# Authenticate (for AI features)
auggie login

# Run the example (from the context directory)
cd examples/python-sdk/context
python -m filesystem_context

# Or run directly
python filesystem_context/main.py
```

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

### Run Examples Directly with tsx

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Executes TypeScript examples directly using tsx without installing project dependencies first. Note: dependencies will be downloaded on each run.

```bash
npx tsx direct-context/index.ts
```

```bash
npx tsx filesystem-context/index.ts
```

```bash
npx tsx file-search-server/index.ts .
```

```bash
npx tsx prompt-enhancer-server/index.ts .
```

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

### Run FileSystem Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/filesystem-context/README.md

Execute the filesystem context example script using tsx. This script demonstrates local directory search and AI-powered Q&A.

```bash
npx tsx examples/context/filesystem-context/index.ts
```

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

### Install GitHub Action Indexer

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/github-action-indexer/README.md

Use the npx command to install the indexer in your repository, which automatically sets up the workflow file, source files, and TypeScript configuration.

```bash
# Install in current directory
cd /path/to/your/repository
npx @augment-samples/github-action-indexer install

# Or specify the repository path
npx @augment-samples/github-action-indexer install /path/to/your/repository
```

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

### Install Auggie SDK

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Install the Auggie Python SDK using pip.

```bash
pip install auggie-sdk
```

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

### Auggie CLI Output Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/github-action-indexer/README.md

Example output from the 'auggie token print' command, showing the format of the access token and tenant URL.

```json
TOKEN={"accessToken":"your-token-here","tenantURL":"https://your-tenant.api.augmentcode.com/","scopes":["read","write"]}
```

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

### Run All Examples with Shell Script

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/user_examples/README.md

Execute all provided Python example scripts using a shell script. This is a convenient way to test multiple functionalities.

```bash
./run_all.sh
```

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

### Run FileSystem Context Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Executes the FileSystem Context example for local directory search via MCP protocol. Ensure large directories are ignored to prevent timeouts.

```bash
python -m filesystem_context
```

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

### Install Auggie CLI

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/filesystem_context/README.md

Install the Auggie CLI tool. This command is used to verify the installation.

```bash
auggie --version
```

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

### Install ACP Adapter

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

Install the Node.js ACP adapter globally using npm.

```bash
npm install -g @zed-industries/claude-code-acp
```

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

### Install and Authenticate Auggie CLI

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/file-search-server/README.md

Install the Auggie CLI globally and authenticate your account. This is a prerequisite for using the file search server.

```bash
npm install -g @augmentcode/auggie@latest
auggie login
```

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

### Install All Augment Commands

Source: https://github.com/augmentcode/auggie/blob/main/examples/commands/README.md

Installs all available Augment command templates from the GitHub repository into the local `~/.augment/commands` directory.

```bash
mkdir -p ~/.augment/commands
cd ~/.augment/commands
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/code-review.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/documentation.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/tests.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/bug-fix.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/performance-optimization.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/security-review.md
```

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

### Setup and Search with GitHub Action Indexer

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/github_action_indexer/README.md

Steps to set up credentials and search an indexed repository locally using the CLI. Ensure you have unzipped the artifact and set the necessary environment variables.

```bash
unzip ~/Downloads/index-state.zip -d .augment-index-state

export AUGMENT_API_TOKEN="your-token"
export AUGMENT_API_URL="https://your-tenant.api.augmentcode.com/"
export BRANCH="main"  # or whatever branch you indexed

python -m github_action_indexer search "authentication functions"
```

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

### Sequential Workflow Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Demonstrates a sequential workflow using agent sessions, where each step builds upon the previous one.

```python
with agent.session() as session:
    session("Step 1")
    session("Step 2 that builds on step 1")
    session("Step 3 that uses both")
```

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

### Start Prompt Enhancer Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/prompt_enhancer_server/README.md

Start the prompt enhancer server by navigating to the context directory and running the Python module. Provide the workspace directory as an argument.

```bash
cd examples/python-sdk/context
python -m prompt_enhancer_server .
```

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

### Run ACP Example Usage

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Execute the acp_example_usage.py script to demonstrate the AuggieACPClient for persistent sessions with the Augment CLI.

```bash
python acp_example_usage.py
```

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

### Run Session Usage Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Execute the session_usage.py script to demonstrate session management for conversation continuity.

```bash
python session_usage.py
```

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

### Install Specific Augment Commands

Source: https://github.com/augmentcode/auggie/blob/main/examples/commands/README.md

Installs selected Augment command templates from the GitHub repository into the local `~/.augment/commands` directory.

```bash
mkdir -p ~/.augment/commands
cd ~/.augment/commands
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/code-review.md
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/commands/documentation.md
```

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

### Install Auggie CLI

Source: https://github.com/augmentcode/auggie/blob/main/README.md

Install the Auggie CLI globally using npm. Node.js version 22 or higher is required.

```sh
npm install -g @augmentcode/auggie@latest
```

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

### Install All Workflows (Bash)

Source: https://github.com/augmentcode/auggie/blob/main/examples/workflows/README.md

Installs all available Augment AI GitHub Actions workflows by downloading their YAML files. Ensure you are in the correct directory before running.

```bash
mkdir -p .github/workflows
cd .github/workflows
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/describe-pr.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/pr-review.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/on-demand-description.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/on-demand-review.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/release.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/test-action.yml
git add .github/workflows/
git commit -m "Add Augment AI workflows"
git push
```

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

### Complete Agent Event Listener Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/AGENT_EVENT_LISTENER.md

This example demonstrates how to implement all agent event listener methods to capture and process various agent outputs like messages, tool calls, and thoughts.

```python
from auggie_sdk import Auggie
from auggie_sdk.acp import AgentEventListener

class MyListener(AgentEventListener):
    def on_agent_message_chunk(self, text: str) -> None:
        print(f"[CHUNK] {text}", end="", flush=True)

    def on_agent_message(self, message: str) -> None:
        print(f"\n[COMPLETE MESSAGE] {message}")

    def on_tool_call(self, tool_call_id: str, title: str,
                    kind: str = None, status: str = None) -> None:
        print(f"\n[TOOL CALL] {title} (kind={kind}, status={status})")

    def on_tool_response(self, tool_call_id: str,
                        status: str = None, content: Any = None) -> None:
        print(f"[TOOL RESPONSE] status={status}")

    def on_agent_thought(self, text: str) -> None:
        print(f"[THINKING] {text}")

# Use the listener
listener = MyListener()
agent = Auggie(listener=listener)

response = agent.run("Read the file test.py and count the lines")
```

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

### Run List PRs Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Execute the list_prs.py script to demonstrate working with GitHub Pull Requests using the SDK.

```bash
python list_prs.py
```

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

### Function Calling Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Demonstrates how the agent can call predefined Python functions to perform specific tasks, returning structured results.

```python
def run_tests(test_file: str) -> dict:
    """
    Run tests from a test file.

    Args:
        test_file: Path to the test file
    """
    # Your test logic
    return {"passed": 10, "failed": 2, "file": test_file}

# Agent can call the function as needed
result = agent.run(
    "Run tests in test_auth.py and analyze the results",
    return_type=dict,
    functions=[run_tests]
)
```

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

### Troubleshoot: Adapter not found

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

If you encounter a 'RuntimeError: npx not found', ensure Node.js is installed from nodejs.org.

```text
RuntimeError: npx not found
```

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

### Searching the Index with CLI

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/github_action_indexer/README.md

Demonstrates how to search the indexed repository using the command-line interface after navigating to the context examples directory. You can search for specific functionalities, patterns, or implementations.

```bash
# From the context examples directory
cd examples/python-sdk/context

# Search for specific functionality
python -m github_action_indexer search "authentication functions"

# Search for error handling patterns
python -m github_action_indexer search "error handling"

# Search for specific implementations
python -m github_action_indexer search "database queries"
```

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

### Example Search Output

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/github-action-indexer/README.md

This is an example of the output format when searching the index, showing the search query, index status, and matching code chunks with file paths and line numbers.

```text
Searching for: "authentication functions"

Loading index state...
Loaded index: 42 files indexed
Last indexed commit: abc123def456
Branch: main

Found 3 result(s):

📄 src/auth/login.ts
   Lines 15-28
   ────────────────────────────────────────────────────────────
     15 │ export async function authenticateUser(
     16 │   username: string,
     17 │   password: string
     18 │ ): Promise<User> {
     19 │   // Authentication logic...
     20 │ }
```

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

### Install Specific Workflows (Bash)

Source: https://github.com/augmentcode/auggie/blob/main/examples/workflows/README.md

Installs selected Augment AI GitHub Actions workflows by downloading their YAML files. This is useful if you only need certain functionalities.

```bash
mkdir -p .github/workflows
cd .github/workflows
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/describe-pr.yml
curl -O https://raw.githubusercontent.com/augmentcode/auggie/main/examples/workflows/pr-review.yml
git add .github/workflows/
git commit -m "Add Augment AI workflows"
git push
```

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

### Start Prompt Enhancer Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/prompt-enhancer-server/README.md

Start the prompt enhancer server using npx and tsx, providing the workspace directory as an argument. This command initiates the server that will process and enhance prompts.

```bash
npx tsx examples/context/prompt-enhancer-server/index.ts .
```

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

### Run the File Search Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/file-search-server/README.md

Execute the file search server example using Node.js with tsx. Provide the directory to search as an argument.

```bash
npx tsx examples/context/file-search-server/index.ts .
```

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

### Conditional Logic Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Illustrates conditional logic based on an agent's response, demonstrating how to handle different scenarios.

```python
exists = agent("Does file exist?", bool)
if exists:
    agent("Process existing file")
else:
    agent("Create new file")
```

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

### Session Continuity Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

Demonstrates how messages sent to the same client instance maintain conversational context.

```python
client = ClaudeCodeACPClient()
client.start()

# First message
client.send_message("My favorite color is blue")

# Second message - remembers the context
response = client.send_message("What is my favorite color?")
print(response)  # Will mention "blue"

client.stop()
```

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

### Login to Auggie

Source: https://github.com/augmentcode/auggie/blob/main/README.md

Authenticate with Auggie to use its services. This command should be run after installation.

```sh
auggie login
```

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

### Initialize and Use AuggieACPClient

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/ARCHITECTURE.md

Demonstrates initializing AuggieACPClient with model, workspace, and listener, then starting it and sending a message. This client directly implements the ACP protocol.

```python
from auggie_sdk.acp import AuggieACPClient

client = AuggieACPClient(
    model="claude-3-5-sonnet-latest",      # CLI arg
    workspace_root="/path/to/workspace",   # CLI arg
    listener=MyEventListener()             # ACP feature
)

client.start()                             # ACP: initialize + newSession
response = client.send_message("...")      # ACP: prompt → returns str
client.clear_context()                     # ACP: clear
client.stop()                              # Cleanup
```

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

### Local Testing Setup and Execution

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/github-action-indexer/README.md

Set environment variables to configure the AugmentCode indexer for local testing and then run the indexing and search commands using npm.

```bash
# Set up environment
export AUGMENT_API_TOKEN="your-token"
export AUGMENT_API_URL="https://your-tenant.api.augmentcode.com/"
export GITHUB_TOKEN="your-github-token"
export GITHUB_REPOSITORY="owner/repo"
export GITHUB_SHA="$(git rev-parse HEAD)"

# Run indexing
npm run index

# Test search
npm run search "your search query"
```

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

### Execute Augment Commands

Source: https://github.com/augmentcode/auggie/blob/main/examples/commands/README.md

Demonstrates how to execute installed Augment commands using the `auggie` CLI tool, specifying the command and the target file or directory.

```bash
auggie "/code-review src/main.py"
auggie "/documentation src/utils.py"
auggie "/tests src/api/"
```

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

### Agent Thinking Process Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/AGENT_EVENT_LISTENER.md

Shows how to capture the agent's internal reasoning before it takes action. Useful for understanding decision-making and debugging.

```text
on_agent_thought("I need to first read the README.md file using the view tool")
  on_agent_thought("Then I'll analyze the content and create a summary")
```

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

### Error Handling Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

Implement try-except blocks to catch and handle potential errors during client operation.

```python
from auggie_sdk.acp import ClaudeCodeACPClient

try:
    client = ClaudeCodeACPClient()
    client.start()
    response = client.send_message("Hello!")
    client.stop()
except ValueError as e:
    print(f"Configuration error: {e}")
except RuntimeError as e:
    print(f"Runtime error: {e}")
except TimeoutError as e:
    print(f"Timeout: {e}")
```

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

### Error Handling Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Illustrates how to handle potential errors during agent execution using try-except blocks for specific Auggie SDK exceptions.

```python
from auggie_sdk.exceptions import AugmentCLIError, AugmentParseError

try:
    result = agent("Complex task", int)
except AugmentParseError:
    print("Could not parse result")
except AugmentCLIError:
    print("Agent execution failed")
```

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

### Iteration with Structured Data Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Shows how to iterate over structured data returned by the agent, processing items based on their properties.

```python
@dataclass
class FileInfo:
    path: str
    size: int

files = agent("Analyze files", list[FileInfo])
for file in files:
    if file.size > 1000:
        agent(f"Optimize {file.path}")
```

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

### Run File Search Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Starts the File Search Server, a REST API for semantic file search with AI summarization. Requires Auggie CLI.

```bash
npm run file-search-server [workspace-directory]
```

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

### Agent Library Simple One-off Request

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/SESSION_CONTINUITY.md

Example of using the Agent Library for a simple, one-off request. This is suitable when session continuity or real-time streaming is not required.

```python
from auggie_sdk import Auggie

agent = Auggie()
result = agent.run("What is 2 + 2?", return_type=int)
print(result)
```

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

### Run Prompt Enhancer Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Starts the Prompt Enhancer Server, an HTTP server that enhances prompts with codebase context. Requires Auggie CLI.

```bash
npm run prompt-enhancer-server [workspace-directory]
```

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

### Generate Security Analysis Workflow

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

Converts a prompt for security analysis into a Python SDK program. This example demonstrates sequential stages, conditional logic, and iteration.

```python
#!/usr/bin/env python3
"""
Security analysis workflow for Python files.
"""

from auggie_sdk import Auggie
from dataclasses import dataclass
from typing import List

@dataclass
class SecurityIssue:
    file: str
    severity: str
    description: str
    line_number: int

def main():
    agent = Auggie(workspace_root=".", model="claude-3-5-sonnet-latest")

    # Stage 1: Get list of Python files
    files = agent("List all Python files in src/ directory", list[str])
    print(f"Found {len(files)} Python files to analyze")

    # Stage 2: Analyze each file for security issues
    all_issues: List[SecurityIssue] = []
    with agent.session() as session:
        for file in files:
            issues = session(
                f"Analyze {file} for security vulnerabilities. "
                f"Return a list of issues found.",
                list[SecurityIssue]
            )
            all_issues.extend(issues)

    print(f"Found {len(all_issues)} total security issues")

    # Stage 3: Create report
    agent(f"Create a security report in security_report.md summarizing "
          f"the {len(all_issues)} issues found across {len(files)} files")

    # Stage 4: Generate fixes for critical issues
    critical_issues = [i for i in all_issues if i.severity == "critical"]

    if critical_issues:
        print(f"Found {len(critical_issues)} critical issues - generating fixes")
        with agent.session() as session:
            for issue in critical_issues:
                session(
                    f"Create a fix for the critical security issue in {issue.file} "
                    f"at line {issue.line_number}: {issue.description}"
                )
    else:
        print("No critical issues found!")

if __name__ == "__main__":
    main()

```

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

### Install Augment Indexer into Target Repository

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/github_action_indexer/README.md

Use this command to copy the Augment indexer code and necessary configuration files into your target repository for production use. This includes the indexer source, workflow file, and updates to `.gitignore`.

```bash
# From the auggie repo, install into your target repository
cd examples/python-sdk/context
python -m github_action_indexer install /path/to/your/repo
```

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

### Tool Call Notification Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/AGENT_EVENT_LISTENER.md

Demonstrates receiving a notification when the agent initiates a tool call. This includes details like the tool's ID, title, kind, and status.

```python
on_tool_call(
    tool_call_id="tc_001",
    title="view",
    kind="read",
    status="pending"
  )
```

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

### Agent Message Chunking Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/AGENT_EVENT_LISTENER.md

Illustrates how the agent streams response messages in chunks. Use this to display partial responses as they arrive, similar to live typing.

```text
on_agent_message_chunk("The ")
  on_agent_message_chunk("answer ")
  on_agent_message_chunk("is ")
  on_agent_message_chunk("4.")
```

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

### Complete Agent Message Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/AGENT_EVENT_LISTENER.md

Shows receiving the agent's full response message after all chunks have been processed. Use this for logging or processing the entire output.

```text
on_agent_message("The answer is 4.")
```

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

### Run Auggie in Project Directory

Source: https://github.com/augmentcode/auggie/blob/main/README.md

Execute Auggie within your project directory. An optional initial prompt can be provided to guide Auggie's actions.

```sh
cd /path/to/your/project
auggie "optional initial prompt"
```

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

### Automatic Session Continuity Example

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/ARCHITECTURE.md

Demonstrates how the agent automatically remembers context across multiple runs without explicit session management. This is a key benefit of the long-running session model.

```python
agent.run("Create a function")
agent.run("Test it")  # Automatically remembers context
```

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

### Adding Function Calling to Generated Code

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

Illustrates how to add custom functions to be used by the agent within a generated SDK program, similar to the initial function calling example.

```python
def run_tests(file: str) -> dict:
    """Run tests for a file."""
    # Your implementation
    return {"passed": 10, "failed": 0}

result = agent.run(
    "Run tests and analyze results",
    return_type=dict,
    functions=[run_tests]
)
```

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

### Prompt Example: Code Refactoring Pipeline

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

An example prompt for refactoring long functions in a codebase, including running tests after refactoring and logging changes.

```plaintext
Find all functions in the codebase that are longer than 50 lines. For
each one, analyze if it can be broken down into smaller functions. If
yes, refactor it into multiple well-named functions with clear
responsibilities. After each refactoring, run the existing tests to
ensure nothing broke. Keep a log of all refactorings performed.
```

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

### Command-Line Usage

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

Illustrates the command-line interface for the prompt_to_code.py script, showing available arguments for help, output path, model selection, and workspace.

```bash
usage: prompt_to_code.py [-h] [-o OUTPUT] [-m MODEL] [-w WORKSPACE] prompt_file

Convert a complex prompt into an Augment SDK program

positional arguments:
  prompt_file           Path to the file containing the prompt to convert

optional arguments:
  -h, --help            show this help message and exit
  -o OUTPUT, --output OUTPUT
                        Output file path for the generated program (default: auto-generated)
  -m MODEL, --model MODEL
                        Model to use for conversion (default: claude-3-5-sonnet-latest)
  -w WORKSPACE, --workspace WORKSPACE
                        Workspace root directory (default: current directory)
```

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

### Sequential Stages with Sessions

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

Demonstrates how to use `agent.session()` to manage sequential steps where each step builds upon the previous one.

```python
with agent.session() as session:
    session("Step 1: Create the base structure")
    session("Step 2: Add features to what we just created")
    session("Step 3: Test everything we built")
```

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

### Clear Context

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

Resets the agent's conversational context, starting a new session.

```python
client.clear_context()  # Restarts the agent with a new session
```

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

### Client Usage with Context Manager

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/CLAUDE_CODE_CLIENT.md

Utilize the ClaudeCodeACPClient as a context manager for automatic start and stop operations.

```python
from auggie_sdk.acp import ClaudeCodeACPClient

with ClaudeCodeACPClient() as client:
    response = client.send_message("Write a hello world function in Python")
    print(response)
```

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

### Index and Search Your Own Repository

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/github_action_indexer/README.md

Adapt the indexing and search commands for your own GitHub repository. Replace placeholders with your repository details and desired search queries.

```bash
# Set to any repo you have access to
export GITHUB_REPOSITORY="your-username/your-repo"
export GITHUB_SHA="main"  # or a specific commit SHA

python -m github_action_indexer index
python -m github_action_indexer search "your search query"
```

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

### Simple Script Template with Auggie SDK

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/README.md

Use this template for basic Auggie agent interactions. It shows how to initialize the agent and perform a single instruction.

```python
#!/usr/bin/env python3
"""
Description of what this script does.
"""

from auggie_sdk import Auggie

def main():
    # Create agent
    agent = Auggie(workspace_root=".", model="claude-3-5-sonnet-latest")

    # Your workflow here
    result = agent("Your instruction")
    print(f"Result: {result}")

if __name__ == "__main__":
    main()

```

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

### Get Staged Git Changes

Source: https://github.com/augmentcode/auggie/blob/main/plugin_marketplace/code-review/agents/local-analyzer.md

Shows changes that have been staged in the index but not yet committed, relative to the current HEAD.

```bash
git diff --staged
```

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

### Get Full Git Diff Against Origin/Main

Source: https://github.com/augmentcode/auggie/blob/main/plugin_marketplace/code-review/agents/local-analyzer.md

Retrieves the complete diff of committed changes between the current HEAD and origin/main.

```bash
git diff origin/main...HEAD
```

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

### Agent Library Session Usage

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/SESSION_CONTINUITY.md

Demonstrates how to use sessions with the Agent Library. Remember to use the session context manager to maintain state across calls.

```python
with agent.session() as session:
    session.run("Step 1")
    session.run("Step 2")
```

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

### Get Unstaged Git Changes

Source: https://github.com/augmentcode/auggie/blob/main/plugin_marketplace/code-review/agents/local-analyzer.md

Displays unstaged changes in the working tree compared to the index, representing uncommitted work-in-progress.

```bash
git diff
```

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

### Health Check API Endpoint

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/prompt_enhancer_server/README.md

Send a GET request to the /health endpoint to check if the prompt enhancer server is running and responsive.

```bash
curl "http://localhost:3001/health"
```

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

### Exclude Directories in .gitignore/.augmentignore

Source: https://github.com/augmentcode/auggie/blob/main/examples/typescript-sdk/context/README.md

Example content for a .gitignore or .augmentignore file to exclude large directories during indexing, preventing timeouts in FileSystem Context.

```bash
# .gitignore or .augmentignore
node_modules/
dist/
*.log
.DS_Store
```

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

### Run File Search Server

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/file_search_server/README.md

Execute the file search server from the context directory or directly.

```bash
# From the context directory
cd examples/python-sdk/context
python -m file_search_server .

# Or run directly
python file_search_server/main.py .
```

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

### Prompt to Code with Custom Workspace

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/docs/PROMPT_TO_CODE.md

Specify a custom workspace directory for the conversion process.

```bash
python prompt_to_code.py my_prompt.txt --workspace /path/to/workspace
```

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

### Ignore Files in FileSystem Context

Source: https://github.com/augmentcode/auggie/blob/main/examples/python-sdk/context/README.md

Example of lines to add to a .gitignore or .augmentignore file to exclude directories and files from indexing in the FileSystem Context.

```bash
# .gitignore or .augmentignore
node_modules/
dist/
*.log
.DS_Store
__pycache__/
*.pyc
```