Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.runaegis.co/llms.txt

Use this file to discover all available pages before exploring further.

Aegis gives you visibility and control over every action your AI agents take. When a coding agent calls a tool — pushing code, creating branches, merging pull requests — Aegis intercepts the request, evaluates it against your configured policies, and either allows it, rewrites it to a safer path, blocks it, or routes it to a human for approval. Every decision is logged in an immutable audit trail.

Quick Start

Connect your first agent to Aegis in under five minutes.

How It Works

Understand how Aegis intercepts and governs agent actions.

Policies

Configure the rules that evaluate every agent tool call.

Integrations

Connect Claude Code, Cursor, or VS Code Copilot via MCP.

Approvals

Review and approve sensitive agent actions in real time.

Audit Trail

Browse the complete log of every agent decision.

​
Get started in four steps

1

Create your account

Sign up at app.runaegis.co with your email, Google, or GitHub account.
2

Connect GitHub

Provide your GitHub username, user ID, and a personal access token with repo, read:user, and workflow scopes. Aegis uses this to inspect repositories and manage pull requests on your behalf.
3

Create a room and connect your agent

Create a room for the repository you want to govern, then paste the generated MCP config snippet into your AI coding tool. Aegis will start intercepting tool calls immediately.
4

Configure your policies

Enable or disable the governance rules that matter most to your team — from protected branch denial to secret detection. Every policy takes effect instantly.
Aegis connects to your agent via the Model Context Protocol (MCP) over Server-Sent Events (SSE). No code changes are required in your agent or repositories.