### Install and Run Agent with Go

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Install dependencies and run the agent using Go commands. Alternatively, use the 'task run' command if Task is configured.

```bash
# Install dependencies
go mod download

# Run the agent
go run main.go

# Or use Task
task run
```

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

### Install Documentation-Agent CLI

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Add the documentation-agent to your Inference Gateway CLI. This command registers the agent and starts it.

```bash
infer agents add documentation-agent http://localhost:8080 \
  --oci ghcr.io/inference-gateway/documentation-agent:latest \
  --run
```

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

### Start Agent Service Individually

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Start only the Documentation Agent service using Docker Compose. This is useful for testing or debugging the agent in isolation.

```bash
docker compose up agent
```

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

### Copy Environment Files

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Copy example environment files to your local directory for configuration. These files are essential for setting up API keys and service parameters.

```bash
cp .env.agent.example .env.agent
cp .env.gateway.example .env.gateway
```

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

### Start Services with Docker Compose

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Launch all required services, including the Documentation Agent and Inference Gateway, using Docker Compose. This command builds the images if necessary and starts the containers.

```bash
docker compose up --build
```

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

### Start Inference Gateway Service Individually

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Start only the Inference Gateway service using Docker Compose. This allows for focused testing or debugging of the gateway component.

```bash
docker compose up inference-gateway
```

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

### Basic Documentation Query

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Perform a basic documentation search query using the A2A debugger. This example demonstrates how to ask general questions about libraries or frameworks.

```bash
docker compose run --rm a2a-debugger tasks submit-streaming "How do I use React hooks?"
```

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

### Get Library Docs Tool

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Fetches up-to-date documentation for a library using its Context7-compatible library ID.

```APIDOC
## Tool: get_library_docs

### Description
Fetches up-to-date documentation for a library using Context7-compatible library ID

### Tags
docs, libraries

### Input Schema
Defined in agent configuration

### Output Schema
Defined in agent configuration
```

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

### Get Agent Information via Curl

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Retrieve agent metadata and capabilities by sending a GET request to the agent's well-known agent card endpoint.

```bash
# Get agent information
curl http://localhost:8080/.well-known/agent-card.json
```

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

### Get Specific Task Details via Debugger

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Retrieve detailed information about a specific task using its ID with the A2A Debugger. Replace `<task ID>` with the actual task identifier.

```bash
docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks get <task ID>
```

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

### Build and Run Agent with Docker

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Build the Docker image for the documentation-agent and run it as a container, mapping the agent's port to the host.

```bash
# Build image
docker build -t documentation-agent .

# Run container
docker run -p 8080:8080 documentation-agent
```

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

### Run Documentation-Agent

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Instructions for running the agent locally using Go or Docker. Ensure the agent is accessible on port 8080.

```bash
# Run the agent
go run .

# Or with Docker
docker build -t documentation-agent .
docker run -p 8080:8080 documentation-agent
```

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

### Build Docker Image with Custom Version

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Build the Docker image for the documentation agent with custom version information, agent name, and description using build arguments.

```bash
# Build with custom version information
docker build \
  --build-arg VERSION=1.2.3 \
  --build-arg AGENT_NAME="My Custom Agent" \
  --build-arg AGENT_DESCRIPTION="Custom agent description" \
  -t documentation-agent:1.2.3 .
```

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

### Project Structure Overview

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Illustrates the directory layout of the inference gateway agent project, highlighting the purpose of main files and directories like server entry point, tools, skills, and configuration.

```bash
. 
├── main.go                       # Server entry point
├── tools/                        # Function-call tools
│   └── read.go                   # Read a file from disk. Returns its contents, optionally sliced by line offset/limit. Use this to load SKILL.md bodies on demand.
│   └── resolve_library_id.go     # Resolves library name to Context7-compatible library ID and returns matching libraries
│   └── get_library_docs.go       # Fetches up-to-date documentation for a library using Context7-compatible library ID
├── skills/                       # Skill directories (SKILL.md + optional assets)
│   └── library-documentation-lookup/# Use this when you need up-to-date documentation for a third-party library or framework before writing code against it. First resolves the library name to a Context7-compatible ID via resolve_library_id when the caller does not already know it (format '/org/project' or '/org/project/version'), then fetches focused, topic-scoped documentation via get_library_docs. Good for filling in unknowns about specific APIs, hooks, configuration options, or version-specific behavior.
│       └── SKILL.md              # Playbook prepended to the system prompt
├── .well-known/                  # Agent configuration
│   └── agent-card.json           # Agent metadata
├── go.mod                        # Go module definition
└── README.md                     # Project documentation

```

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

### Build Docker Image with Default Version

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Build the Docker image for the documentation agent using default version information specified in the ADL.

```bash
# Build with default values from ADL
docker build -t documentation-agent .
```

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

### List Available Tools

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

View all the tools and functionalities exposed by the Documentation Agent via the A2A debugger. This helps in understanding the agent's capabilities.

```bash
docker compose run --rm a2a-debugger tools list
```

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

### Development Commands for documentation-agent

Source: https://github.com/inference-gateway/documentation-agent/blob/main/CLAUDE.md

Common tasks for managing the documentation-agent project, including code generation, running the server, testing, linting, formatting, building, and cleaning artifacts.

```bash
# Generate/regenerate code from ADL specification
task generate

# Run the agent in development mode (debug enabled, port 8080)
task run

# Run tests (note: no tests currently exist)
task test
task test:cover  # with coverage

# Code quality
task lint         # Run golangci-lint
task fmt          # Format code with go fmt

# Build
task build        # Creates bin/documentation-agent
task docker:build # Build Docker image

# Clean build artifacts
task clean
```

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

### Build Agent Docker Image

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Build only the Docker image for the Documentation Agent. This is useful for development when only the agent needs to be updated.

```bash
docker compose build agent
```

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

### Testing Commands for Inference Gateway Agent

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Provides commands to run tests for the agent, including running all tests and running tests with coverage reporting.

```bash
# Run tests
task test
go test ./...

```

```bash
# Run with coverage
task test:coverage

```

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

### Run A2A Debugger Help Command

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Execute the help command for the A2A debugger to see available options.

```bash
docker compose run --rm a2a-debugger --help
```

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

### Testing Individual Components with Go and A2A Debugger

Source: https://github.com/inference-gateway/documentation-agent/blob/main/CLAUDE.md

Commands for running specific Go test files and debugging the agent using the A2A Debugger. Ensure tests are added before using the `go test` command.

```bash
# Run specific test file (when tests are added)
go test -v ./path/to/package -run TestFunctionName

# Debug with A2A Debugger
docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest \
  --server-url http://localhost:8080 tasks submit "Your query"
```

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

### Configure Direct OpenAI Connection

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Modify the agent's environment variables to connect directly to an OpenAI-compatible API, bypassing the Inference Gateway. This requires updating the base URL, provider, and model.

```bash
# Example: Direct OpenAI connection
A2A_AGENT_CLIENT_BASE_URL=https://api.openai.com/v1
A2A_AGENT_CLIENT_PROVIDER=openai
A2A_AGENT_CLIENT_MODEL=gpt-4
```

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

### Development Tasks for Documentation Agent

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Common development tasks for the Documentation Agent, including code generation, testing, building, linting, and formatting.

```bash
# Generate code from ADL
task generate
```

```bash
# Run tests
task test
```

```bash
# Build the application
task build
```

```bash
# Run linter
task lint
```

```bash
# Format code
task fmt
```

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

### Submit Streaming Task with A2A Debugger

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Submit a streaming task to the Documentation Agent using the A2A debugger CLI. This is useful for interactive queries where results are received incrementally.

```bash
docker compose run --rm a2a-debugger tasks submit-streaming "What's the latest version of NextJS?"
```

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

### Read File Tool

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Reads a file from disk, optionally sliced by line offset/limit.

```APIDOC
## Tool: Read (built-in)

### Description
Read a file from disk. Returns its contents, optionally sliced by line offset/limit. Use this to load SKILL.md bodies on demand.

### Parameters
- **file_path** (string) - Required - The path to the file.
- **offset** (integer) - Optional - The starting line number.
- **limit** (integer) - Optional - The maximum number of lines to return.
```

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

### List Tasks on A2A Agent via Debugger

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Use the A2A Debugger to list all tasks currently managed by your agent. This is useful for monitoring and tracking task status.

```bash
docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks list
```

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

### Submit a Task to A2A Agent via Debugger

Source: https://github.com/inference-gateway/documentation-agent/blob/main/README.md

Use the A2A Debugger to submit a task to your agent. Ensure the agent is running and accessible at the specified server URL.

```bash
docker run --rm -it --network host ghcr.io/inference-gateway/a2a-debugger:latest --server-url http://localhost:8080 tasks submit "What are your skills?"
```

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

### Agent Metadata and Capabilities

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Retrieve metadata and capabilities of the agent.

```APIDOC
## GET /.well-known/agent-card.json

### Description
Agent metadata and capabilities

### Method
GET

### Endpoint
/.well-known/agent-card.json
```

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

### Library-Specific Search

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Execute a targeted search within the documentation of a specific library using the A2A debugger. This allows for precise information retrieval.

```bash
docker compose run --rm a2a-debugger tasks submit-streaming "Search Vue.js documentation for composition API"
```

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

### Build Docker Images with No Cache

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Build Docker images without using the cache. This ensures that all layers are rebuilt, which can be helpful for troubleshooting build issues.

```bash
docker compose build --no-cache
```

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

### Submit Non-Streaming Task

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Submit a non-streaming task to the Documentation Agent. This is suitable for queries where the entire response is expected at once.

```bash
docker compose run --rm a2a-debugger tasks submit "What is TypeScript?"
```

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

### Stop All Docker Compose Services

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

Use this command to gracefully stop all services defined in your Docker Compose file.

```bash
docker compose down
```

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

### Resolve Library ID Tool

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Resolves a library name to its Context7-compatible library ID.

```APIDOC
## Tool: resolve_library_id

### Description
Resolves library name to Context7-compatible library ID and returns matching libraries.

### Tags
docs, libraries

### Input Schema
Defined in agent configuration

### Output Schema
Defined in agent configuration
```

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

### Stop Docker Compose Services and Remove Volumes

Source: https://github.com/inference-gateway/documentation-agent/blob/main/example/README.md

This command stops all services and removes associated Docker volumes, ensuring a clean state.

```bash
docker compose down -v
```

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

### Health Check

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Check the health status of the agent.

```APIDOC
## GET /health

### Description
Health check endpoint

### Method
GET

### Endpoint
/health
```

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

### A2A Operations

Source: https://github.com/inference-gateway/documentation-agent/blob/main/AGENTS.md

Perform A2A operations including skill execution and streaming.

```APIDOC
## POST /a2a

### Description
JSON-RPC endpoint for all A2A operations (skill execution, streaming, etc.)

### Method
POST

### Endpoint
/a2a

### Request Body
- **method** (string) - Required - The A2A method to call.
- **params** (object) - Optional - Parameters for the A2A method.

### Response
#### Success Response (200)
- **result** (any) - The result of the A2A operation.
- **error** (object) - An error object if the operation failed.
```

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