### Setup and Run Orion Web UI

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/README.md

Navigate to the web UI directory, install its dependencies, and start the development server. The UI will be available at http://localhost:3000.

```bash
cd orion-web
npm install
npm run dev
```

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

### Install Dependencies and Start Orion API Server

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/orion-web/README.md

Installs development dependencies and starts the Orion API server. Ensure you are in the orion-agent root directory.

```bash
pip install -e ".[dev]"

uvicorn orion.api.server:app --reload --port 8001
```

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

### Install Dependencies and Start Web UI Development Server

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/orion-web/README.md

Navigates to the orion-web directory, installs Node.js dependencies, and runs the development server for the Orion Web UI. This is typically done once for dependency installation.

```bash
cd orion-web

npm install

npm run dev
```

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

### CLI Setup Wizard for Google OAuth

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/GOOGLE_SETUP.md

Initiates the CLI wizard to guide through Google Cloud OAuth app creation and saves credentials.

```bash
/google setup
```

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

### Setup Python Virtual Environment and Install Dependencies

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/README.md

Set up a Python virtual environment and install development dependencies for the Orion Agent backend. This includes installing the project in editable mode with development extras.

```bash
cd orion-agent
python -m venv .venv
.venv\Scripts\activate
pip install -e ".[dev]"
```

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

### Install Orion Agent on Windows

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Instructions for installing Orion Agent on Windows, including Python setup, virtual environment activation, and pip installation.

```powershell
python -m venv .venv
.venv\Scripts\activate
pip install orion-agent
```

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

### Example Prompt Injection Context

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/LEARNING_SYSTEM.md

Illustrates how Orion injects relevant success and anti-patterns into the LLM prompt to guide response generation. Anti-patterns are weighted to be avoided.

```text
[Memory Context]
Success patterns for this task type:
- This project uses custom exception classes for error handling
- Always include type hints in function signatures
- Tests follow pytest conventions with fixtures

Anti-patterns to avoid:
- Don't use bare except clauses (rated 1/5 on 2025-01-15)
- Don't modify __init__.py without checking circular imports
```

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

### Start Web Frontend

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/USER_GUIDE.md

Command to start the Orion web frontend development server using npm. Navigate to the 'orion-web' directory before running this command.

```bash
cd orion-web
npm run dev

```

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

### Start Orion Agent Web Frontend

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/FAQ.md

Starts the development server for the Orion Agent Web UI. Navigate to the `orion-web` directory and run this command.

```bash
cd orion-web && npm run dev
```

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

### Start Orion Web UI

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/USER_GUIDE.md

Navigate to the web UI directory and run this command to start the development server for the Orion Agent's web interface.

```bash
cd orion-web
npm run dev
```

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

### Install Orion Agent from source for development

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Clones the repository, sets up a virtual environment, and installs the agent in development mode. Includes development dependencies.

```bash
git clone https://github.com/phoenixlink-cloud/orion-agent.git
cd orion-agent
python -m venv .venv
# Windows:
.venv\Scripts\activate
# macOS/Linux:
source .venv/bin/activate
pip install -e ".[dev]"
```

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

### Install Development Dependencies

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/CONTRIBUTING.md

Install the project in development mode with all specified dependencies. This is typically done before running tests or making code changes.

```bash
pip install -e ".[dev,web,training]"
```

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

### Start Ollama Serve

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/TROUBLESHOOTING.md

Ensure the Ollama service is running to enable local LLM connections. This command starts the Ollama server.

```bash
ollama serve
```

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

### Install Orion Agent from source

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/GETTING_STARTED.md

Clone the repository and install the Orion Agent from source. This is useful for development or if you need the latest unreleased features.

```bash
git clone https://github.com/phoenixlink-cloud/orion-agent.git
cd orion-agent
pip install -e ".[dev]"
```

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

### Install All Orion Agent Integrations

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INTEGRATIONS.md

Use this command to install all available integrations for the Orion Agent. This is useful for setting up the agent with the maximum number of supported services.

```bash
pip install orion-agent[all]
```

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

### Project Memory Example (JSON)

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/MEMORY_SYSTEM.md

An example of a JSON object stored in Project Memory, detailing a learned coding pattern.

```json
{
  "patterns": [
    {
      "id": "auth-error-handling",
      "content": "This project uses custom AuthError exceptions",
      "confidence": 0.85,
      "access_count": 12,
      "created": "2025-02-01T10:30:00Z"
    }
  ]
}
```

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

### Install Orion Agent

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/README.md

Install the Orion Agent using pip. Use the '[all]' extra for full integrations.

```bash
pip install orion-agent
```

```bash
pip install orion-agent[all]
```

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

### Conventional Commit Examples

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/data/seed/skills/git-workflow/SKILL.md

Examples demonstrating the application of conventional commit messages for different types of changes. These examples illustrate how to structure commit messages effectively.

```git
feat(auth): add OAuth2 login flow
```

```git
fix(api): handle null response from payment provider
```

```git
docs(readme): add deployment instructions
```

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

### Start API Server

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/USER_GUIDE.md

Command to start the Orion API server using Uvicorn. Ensure the application entry point and port are correctly specified.

```bash
uvicorn orion.api.server:app --port 8001

```

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

### Example Complete Orion Agent Configuration

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/CONFIGURATION.md

A comprehensive example of the Orion agent's config.yaml file, demonstrating settings for LLM, governance, memory, agents, sandbox, and logging.

```yaml
# ~/.orion/config.yaml
llm:
  provider: openai
  model: gpt-4o
  temperature: 0.3
  max_tokens: 4096

governance:
  mode: pro

memory:
  tier2_retention_days: 30
  promotion_threshold: 0.85
  auto_consolidate: true

agents:
  enable_table_of_three: true

sandbox:
  mode: auto
  timeout: 60
  network: false

logging:
  level: INFO
  structured: false
```

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

### Install Orion Agent

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Install the Orion Agent package using pip. Use the `[all]` extra for full integrations or install from source for development.

```bash
pip install orion-agent
```

```bash
pip install orion-agent[all]
```

```bash
git clone https://github.com/phoenixlink-cloud/orion-agent.git
cd orion-agent
pip install -e ".[dev]"
```

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

### Troubleshoot 'Permission denied' during installation

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Advises using a virtual environment to avoid permission errors when installing Orion Agent, especially when using system Python.

```bash
python -m venv .venv
source .venv/bin/activate  # or .venv\Scripts\activate on Windows
pip install orion-agent
```

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

### Start Orion API Server

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Provides commands to start the Orion API server using uvicorn. Use the --reload flag for development to automatically restart the server on code changes. The Web UI can be started separately in another terminal.

```bash
# Start FastAPI server
uvicorn orion.api.server:app --port 8001
```

```bash
# Start with auto-reload for development
uvicorn orion.api.server:app --port 8001 --reload
```

```bash
# Start Web UI (separate terminal)
cd orion-web
npm install
npm run dev  # http://localhost:3000
```

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

### Position Size Calculation Example

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/data/seed/skills/risk-management-position-sizing/SKILL.md

An example demonstrating the calculation of position size for a specific trade scenario. This helps in understanding how to apply the formula with real trading parameters.

```text
Position Size =  / (50 pips x ) =  /  = 0.20 lots (2 mini lots)
```

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

### Start Orion Agent CLI

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/CLI_REFERENCE.md

Use these commands to start the Orion Agent in interactive mode, with a specific workspace, or in a particular mode. The `--version` and `--help` flags display agent version and usage information, respectively.

```bash
# Start interactive REPL
orion
```

```bash
# Start with specific workspace
orion --workspace /path/to/project
```

```bash
# Start in specific mode
orion --mode pro
```

```bash
# Show version
orion --version
```

```bash
# Show help
orion --help
```

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

### Install Orion Agent with specific integration groups

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Installs Orion Agent with specific sets of integrations. Choose the group that matches your needs.

```bash
pip install orion-agent[voice]
```

```bash
pip install orion-agent[image]
```

```bash
pip install orion-agent[messaging]
```

```bash
pip install orion-agent[dev]
```

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

### Starting Orion Agent

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/CLI_REFERENCE.md

Commands to start the Orion Agent in various modes, including interactive REPL, with a specific workspace, or in a specific mode. Also includes commands to display version and help information.

```APIDOC
## Starting Orion Agent

### Description
Commands to start the Orion Agent in various modes, including interactive REPL, with a specific workspace, or in a specific mode. Also includes commands to display version and help information.

### Commands

- `orion`
  - Starts the Orion Agent in interactive REPL mode.
- `orion --workspace <path>`
  - Starts the Orion Agent with a specific workspace directory.
- `orion --mode <mode>`
  - Starts the Orion Agent in a specific mode (e.g., `pro`).
- `orion --version`
  - Displays the current version of the Orion Agent.
- `orion --help`
  - Displays the help information for the Orion Agent CLI.
```

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

### Start Orion Agent CLI

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Launch the Orion Agent command-line interface. Options include starting an interactive REPL, specifying a workspace, or setting a specific operational mode.

```bash
orion
```

```bash
orion --workspace /path/to/project
```

```bash
orion --mode pro
```

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

### Get Skill Details

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Fetches detailed information for a specific skill, including full instructions.

```bash
# Get skill with full instructions
curl http://localhost:8001/api/ara/skills/code-review
```

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

### Get Orion Agent Version

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/TROUBLESHOOTING.md

Use the `orion --version` command to retrieve the installed Orion Agent version. This is required when opening a new issue or contacting support.

```bash
orion --version
```

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

### Dockerfile for Orion Agent

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/DEPLOYMENT.md

Defines the Dockerfile for building the Orion Agent image. It installs system dependencies, Python packages, exposes the API port, and sets up a health check and the command to start the API server.

```dockerfile
FROM python:3.11-slim

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*

# Install Orion
COPY . .
RUN pip install --no-cache-dir -e ".[all]"

# Expose API port
EXPOSE 8001

# Health check
HEALTHCHECK --interval=30s --timeout=5s --retries=3 \
  CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:8001/health')"

# Start API server
CMD ["uvicorn", "orion.api.server:app", "--host", "0.0.0.0", "--port", "8001"]
```

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

### Configure Provider and Model Settings via CLI

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Manage LLM provider, model, and generation parameters like temperature using the `/settings` command.

```bash
> /settings key openai sk-your-key-here
API key set for openai
```

```bash
> /settings provider anthropic
Provider set: anthropic
```

```bash
> /settings model claude-3-5-sonnet-20241022
Model set: claude-3-5-sonnet-20241022
```

```bash
> /settings temperature 0.3
Temperature set: 0.3
```

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

### Install C++ Build Tools on Windows

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/TROUBLESHOOTING.md

Install the necessary C++ build tools for Windows if you encounter build errors during installation. This command uses winget to install Visual Studio 2022 Build Tools.

```powershell
winget install Microsoft.VisualStudio.2022.BuildTools
```

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

### Install Orion Agent on macOS

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Instructions for installing Orion Agent on macOS using Homebrew for Python and pip for the agent installation.

```bash
brew install python@3.11
pip3 install orion-agent
```

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

### Install Orion Agent on Linux (Debian/Ubuntu)

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Instructions for installing Orion Agent on Debian/Ubuntu systems, including system package updates and pip installation.

```bash
sudo apt update
sudo apt install python3.11 python3.11-venv python3-pip git
pip3 install orion-agent
```

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

### Initiate Google Account Login via CLI

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/GOOGLE_SETUP.md

Starts the Google OAuth login flow using the CLI command.

```bash
/google login
```

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

### Install Orion Agent via pip

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/GETTING_STARTED.md

Use this command to install the Orion Agent package. Recommended for most users.

```bash
pip install orion-agent
```

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

### User Feedback Example

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/MEMORY_SYSTEM.md

Illustrates how a user provides feedback on Orion's response, which aids in learning and pattern storage.

```text
> Add error handling to the API endpoint

[Orion provides solution]

Rate this response (1-5, or skip): 5
Feedback recorded. Pattern stored.
```

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

### AEGIS Audit Log Example

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/AEGIS.md

Example entries from the AEGIS audit log, showing validation decisions, actions, and reasons.

```log
2025-02-10 14:23:45 | AEGIS | PASS | read | src/main.py | mode=pro
2025-02-10 14:23:52 | AEGIS | PASS | modify | src/main.py | mode=pro | approved=user
2025-02-10 14:24:01 | AEGIS | FAIL | modify | /etc/passwd | reason=path_escape
2025-02-10 14:24:15 | AEGIS | FAIL | run | rm -rf / | reason=dangerous_pattern
```

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

### Manage Skills Programmatically with SkillLibrary

Source: https://context7.com/phoenixlink-cloud/orion-agent/llms.txt

Demonstrates how to initialize, load, list, create, import, and resolve skills using the SkillLibrary. Skills can be managed programmatically for custom AI agent configurations. Ensure the skill path is correct when importing.

```python
from pathlib import Path
from orion.ara.skill_library import SkillLibrary

# Initialize library
library = SkillLibrary()
loaded, warnings = library.load_all()
print(f"Loaded {loaded} skills, {library.group_count} groups")

# List approved skills
skills = library.list_skills(approved_only=True)
for skill in skills:
    print(f"{skill.name}: {skill.description} (trust: {skill.trust_level})")

# Create a new custom skill
skill, scan_result = library.create_skill(
    name="my-deployment",
    description="Deploy to staging environment",
    instructions="""## Deployment Procedure

1. Run tests: `pytest tests/`
2. Build: `npm run build`
3. Deploy: `kubectl apply -f k8s/staging/`
""",
    tags=["devops", "deployment"],
    author="my-team"
)
print(f"Created: {skill.name}, approved: {skill.aegis_approved}")

# Import skill from external directory
imported, scan, warnings = library.import_skill(Path("/path/to/skill-folder"))
if imported:
    print(f"Imported: {imported.name}, trust: {imported.trust_level}")

# Resolve skills for a role
role_skills = library.resolve_skills_for_role(
    assigned_skills=["code-review"],
    assigned_skill_groups=["developer-core"]
)
for s in role_skills:
    print(f"Role has skill: {s.name}")
```

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

### Verify Orion Agent installation

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/INSTALLATION.md

Checks the installed Orion Agent version and runs a diagnostic command to ensure all components are functioning correctly.

```bash
# Check version
orion --version

# Run diagnostics
orion
> /doctor
```

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

### Connect to Orion API

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/CLI_REFERENCE.md

Use `/connect` to establish a connection from the CLI to the Orion API server, displaying the connection endpoint.

```bash
> /connect
Connected to Orion API at http://localhost:8001
```

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

### Display Evolution Summary

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/LEARNING_SYSTEM.md

Use the /evolution command to view a summary of the agent's performance, including key metrics, strengths, weaknesses, and recommendations for improvement.

```bash
> /evolution

Evolution Summary:
  Total tasks: 847
  Approval rate: 82.3%
  Quality trend: Improving (+3.2% this month)
  
  Strengths:
    - Python refactoring (94% approval)
    - Bug fixing (89% approval)
  
  Weaknesses:
    - Test generation (68% approval)
    - Documentation (71% approval)
  
  Recommendations:
    [HIGH] Improve test generation by studying project test patterns
    [MEDIUM] Review documentation style preferences
```

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

### Start and Monitor Docker Compose Services

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/DEPLOYMENT.md

Commands to start the Docker Compose services in detached mode and to follow the logs of the Orion API service.

```bash
docker-compose up -d

docker-compose logs -f orion-api
```

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

### Start Orion Agent API Server

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/README.md

Launch the Orion Agent API server using Uvicorn. The server will be accessible on port 8001.

```bash
uvicorn orion.api.server:app --port 8001
```

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

### Structured Log Example

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/DEPLOYMENT.md

An example of a structured log entry in JSON format, showing timestamp, level, correlation ID, component, event, and detailed information.

```json
{
  "timestamp": "2025-02-10T14:23:45Z",
  "level": "INFO",
  "correlation_id": "abc-123",
  "component": "aegis",
  "event": "validation_pass",
  "details": {
    "action": "modify",
    "path": "src/main.py",
    "mode": "pro"
  }
}
```

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

### Workspace Confinement Examples

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/AEGIS.md

Illustrates file operations that are inside or outside the workspace directory. Operations escaping the workspace or using absolute paths outside are rejected.

```text
/home/user/project/src/main.py     (inside workspace -- PASS)
/home/user/project/../.ssh/id_rsa  (escapes workspace -- FAIL)
/etc/passwd                         (absolute path outside -- FAIL)
```

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

### Configure Ollama for local LLM

Source: https://github.com/phoenixlink-cloud/orion-agent/blob/main/docs/GETTING_STARTED.md

Install Ollama and pull a model like Llama3 for local LLM usage. This option is free and does not require an external API key.

```bash
# Install Ollama from https://ollama.ai
ollama pull llama3
```