Charter — Agent Context Manager
One trusted context for every AI agent on your project.
AI agents can do remarkable work, but a new session doesn’t automatically inherit the decisions, constraints, and lessons from earlier work. Across long-running projects, sessions change, conversations are compacted, and different agents can end up working from different assumptions.
Charter is a small, local, version-controlled home for the decisions that govern a software project. Agents can retrieve that governing context and propose changes to it; you review and approve what becomes authoritative. Charter can provide context and ask agents to use it, but it cannot guarantee an agent will follow it.
The Charter Story
Managing Context in Multi-Agent Development
A five-part story of building software with AI agents, the problem of keeping them aligned, and how Charter approaches it.
Part 1
The Experience
AI Can Build Fast. Keeping Agents Aligned Is the Hard Part.
AI doesn’t just accelerate development. It also accelerates the rate at which your project accumulates decisions, assumptions, changes, and unfinished work.
Read Part 1 →
Part 2
The Problem
The Missing Layer: Governing Context
Simply making information available doesn’t tell an agent which decisions are authoritative.
Read Part 2 →
Part 3
The Approach
Agents Can Propose. People Decide.
I’m simply separating the work of developing a proposal from the authority to approve a decision.
Read Part 3 →
Part 4
The Implementation
How Charter Works
Keep the decisions that govern a project in a durable, organized location, make that information available to AI agents, and provide a controlled way for the context to evolve.
Read Part 4 →
Part 5
The Workflow
Putting Charter to Work
The objective isn’t to document everything. It’s to preserve the decisions that matter and make them available when agents need them.
Read Part 5 →Source code
Charter’s source code is public on GitHub, under the Apache License 2.0.