IntelGraph AI Symphony Aider Config

A sophisticated Aider configuration optimized for agent-centric development workflows with local LLM routing via LiteLLM proxy. Designed for efficient resource usage on M2 16GB systems while supporting multiple specialized agent roles.

What This Skill Does

Configures Aider to work with a local LiteLLM proxy server that routes requests to different models based on role-specific needs. Includes performance optimizations, git integration, and smart file handling patterns for modern development workflows.

Configuration Overview

This `.aider.conf.yaml` file sets up:

Agent Role Quick-Switch

The configuration supports multiple agent personas through model switching:

| Agent | Model Route | Purpose | Requirements |

|-------|-------------|---------|--------------|

| **guy** | `openai/local/llama` | Architect/coder (default) | Local LiteLLM |

| **aegis** | `openai/local/llama-cpu` | Security review | Local LiteLLM |

| **hermes** | `openai/local/llama-small` | Quick CI/PR tasks | Local LiteLLM |

| **elara** | `openai/gemini/1.5-pro` | Research | `GOOGLE_API_KEY` |

| **power** | `openai/xai/grok-code-fast-1` | Power burst coding | `XAI_API_KEY` |

Setup Instructions

1. Install LiteLLM Proxy

```bash

pip install litellm[proxy]

```

2. Create LiteLLM Config

Create a `litellm_config.yaml` file with your model routes:

```yaml

model_list:

- model_name: local/llama

litellm_params:

model: ollama/qwen2.5-coder:14b

api_base: http://localhost:11434

- model_name: local/llama-cpu

litellm_params:

model: ollama/qwen2.5-coder:7b

api_base: http://localhost:11434

- model_name: local/llama-small

litellm_params:

model: ollama/qwen2.5-coder:3b

api_base: http://localhost:11434

general_settings:

master_key: sk-anything

```

3. Start LiteLLM Proxy

```bash

litellm --config litellm_config.yaml --port 4000

```

4. Create Aider Config File

Save the following as `.aider.conf.yaml` in your project root:

```yaml

IntelGraph AI Symphony - Agent-Centric Aider Config

Default to your Qwen route (fast, low-VRAM coder)

model: openai/local/llama

Aider reads these dashed keys (not an "openai:" block)

openai-api-base: http://127.0.0.1:4000/v1

openai-api-key: sk-anything

Quality & performance settings

pretty: true

stream: true

auto-commits: false

git: true

edit-format: diff

show-diffs: true

max-chat-history-tokens: 8192

Performance tuning for M2 16GB

cache-prompts: true

restore-chat-history: false

File handling

encoding: utf-8

ignore-globs: |

*.log

*.tmp

**/node_modules/**

**/dist/**

**/build/**

**/.vite/**

**/.next/**

**/coverage/**

rag/index/*.duckdb*

**/*.pyc

**/__pycache__/**

```

5. Launch Aider

**Default agent (guy - architect/coder):**

```bash

aider

```

**Security review agent (aegis):**

```bash

aider --model openai/local/llama-cpu

```

**Quick CI/PR agent (hermes):**

```bash

aider --model openai/local/llama-small

```

**Research agent (elara) - requires GOOGLE_API_KEY:**

```bash

export GOOGLE_API_KEY=your_key_here

aider --model openai/gemini/1.5-pro

```

**Power burst agent - requires XAI_API_KEY:**

```bash

export XAI_API_KEY=your_key_here

aider --model openai/xai/grok-code-fast-1

```

Key Configuration Options

Quality Settings

Git Integration

Performance Optimization

File Exclusions

Automatically ignores:

Usage Examples

Example 1: Default Coding Session

```bash

Start with default guy agent (Qwen coder)

aider

In Aider chat:

/add src/components/Button.tsx

Refactor this button component to use Tailwind CSS instead of inline styles

```

Example 2: Security Review

```bash

Switch to aegis for security-focused review

aider --model openai/local/llama-cpu

In Aider chat:

/add src/api/auth.ts

Review this authentication code for security vulnerabilities

```

Example 3: Quick PR Description

```bash

Use hermes for fast CI tasks

aider --model openai/local/llama-small

In Aider chat:

/run git diff main

Generate a concise PR description from this diff

```

Customization

Adjust Memory Limits

For systems with more/less RAM:

```yaml

max-chat-history-tokens: 16384 # More context (needs more RAM)

max-chat-history-tokens: 4096 # Less context (saves RAM)

```

Add Custom Model Routes

In your `litellm_config.yaml`:

```yaml

model_list:

- model_name: local/custom-model

litellm_params:

model: ollama/your-model:tag

api_base: http://localhost:11434

```

Then use with:

```bash

aider --model openai/local/custom-model

```

Add Project-Specific Ignores

Extend the `ignore-globs` section:

```yaml

ignore-globs: |

# ... existing patterns ...

**/your-custom-dir/**

*.your-extension

```

Important Notes

Troubleshooting

**Connection Error:**

```bash

Ensure LiteLLM proxy is running

litellm --config litellm_config.yaml --port 4000

```

**Model Not Found:**

```bash

Pull required Ollama model

ollama pull qwen2.5-coder:14b

```

**Out of Memory:**

Related Skills

Search the marketplace for: