# Internal Context Feeds

> Reads across your codebases and environments and emits machine-readable ground truth, so AI agents act on reality, not guesses.

Canonical: https://www.adjective.us/ultimus/environment-context

**Internal Environments & Codebases.** A tear-away module of Ultimus.

- Reads across codebases and environments
- Emits machine-readable ground truth
- Maps architecture, constraints, dependencies

Every agent starts with the same problem. It does not know where it is.

This is the answer. It is the context of your environment, read across every codebase and system it touches. Architecture. Constraints. Dependencies. Structured, and readable by a machine.

Agents act on reality, not guesses.

## What it emits

- The architecture as it exists across your codebases, not as the wiki remembers it.
- The constraints an agent must respect before it touches anything.
- The dependencies that a change will pull on.

## Where it sits

This is the context of the inside. [External context feeds](/ultimus/external-context) are the context of the outside. [Context management](/ultimus/context-management) sits between them, and everything downstream reads the result.

## FAQ

**What do Internal Context Feeds produce?**

Structured, machine-readable ground truth about your architecture, constraints, and dependencies, read across every codebase and environment it touches.

**Why do agents need it?**

Agents cannot operate blindly. Without ground truth they act on guesses. This gives them the context of your environment to act on.

**Can I add it to a deployment I already run?**

Yes. Every module tears away. Add this one to your existing architecture, or take the whole of Ultimus.