IntelGraph Multi-Agent Aider Configuration

Configure Aider to work with a local LiteLLM proxy that routes to multiple specialized AI agents (architect, security reviewer, CI assistant, researcher, power-burst). Optimized for M2 16GB machines with sensible defaults.

What This Configuration Does

Sets up Aider with:

Instructions

1. Create the Aider Configuration File

Create a file named `.aider.conf.yml` in your project root with the following content:

```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__/**

```

2. Set Up LiteLLM Proxy (Required)

This configuration assumes you have a LiteLLM proxy running at `http://127.0.0.1:4000` with these routes configured:

If you don't have LiteLLM set up, install and configure it according to your agent routing needs.

3. Launch Aider with Default Agent

```bash

aider

```

This uses the default `openai/local/llama` route (fast coder).

4. Switch Agents for Specific Tasks

Override the model with `--model` flag to use specialized agents:

**Architect/Coder (default):**

```bash

aider --model openai/local/llama

```

**Security Review (Aegis):**

```bash

aider --model openai/local/llama-cpu

```

**Quick CI/PR Tasks (Hermes):**

```bash

aider --model openai/local/llama-small

```

**Research (Elara, requires GOOGLE_API_KEY):**

```bash

aider --model openai/gemini/1.5-pro

```

**Power Burst (requires XAI_API_KEY):**

```bash

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

```

5. Customize Ignore Patterns

Add project-specific patterns to `ignore-globs` section. Common patterns are already included (node_modules, build artifacts, cache directories).

Configuration Explanation

Key Settings

Agent Roles

| Agent | Model Route | Purpose | Hardware |

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

| Guy | `openai/local/llama` | Architect/Coder | Local Qwen (low VRAM) |

| Aegis | `openai/local/llama-cpu` | Security review | CPU fallback |

| Hermes | `openai/local/llama-small` | Quick CI/PR tasks | Small/fast model |

| Elara | `openai/gemini/1.5-pro` | Research | Cloud (Google API) |

| Power | `openai/xai/grok-code-fast-1` | Heavy lifting | Cloud (xAI) |

Notes

Example Workflow

```bash

Start coding session with default architect agent

aider

Switch to security review agent for audit

aider --model openai/local/llama-cpu src/auth/*.ts

Use research agent for documentation

aider --model openai/gemini/1.5-pro README.md docs/

```