Short answer: there is no single best tool, because “agent config” is several jobs: syncing MCP servers across tools, keeping memory across sessions, local evidence and code structure, and managing dotfiles. Below are the current options grouped by the job, with an honest note on each. Brigade and the Escoffier Labs tools around it are on this list, and so are the tools they do not replace.
If you want to sync MCP servers across tools
- Brigade keeps one MCP catalog and merges it into each tool’s native config (Claude Code, Codex, Cursor, VS Code, OpenCode, Antigravity, Grok, Hermes, OpenClaw), dry-run by default, with a receipt per change. It also handles tools and memory, so this is the more complete, heavier option.
- add-mcp and config-sync scripts copy MCP entries into a few client configs. Lighter, no review gate or receipts, MCP only.
- chezmoi is a general dotfile manager. You can template MCP configs with it, but it does not understand each tool’s shape or show a per-tool diff.
If you want memory that persists across sessions
- mem0 is a memory layer you build into an app through an SDK or hosted API. For products you are developing, not for wiring the CLIs you run.
- Letta (formerly MemGPT) is an agent server with stateful memory blocks, from the MemGPT line of work. A framework more than a drop-in for existing CLIs.
- agentmemory gives coding-agent CLIs persistent memory and handoffs across tools, focused on memory.
- Brigade keeps memory as reviewed markdown in your repo with a slim per-session index, alongside MCP and tools, under one review gate.
- Native harness memory (each tool’s own auto-memory) is the zero-setup option, but it is a per-tool silo that does not cross agents and writes without review.
If you want local structure and evidence (built into Brigade)
- Code intelligence (formerly GraphTrail) is Brigade’s built-in local code graph: callers, callees, impact for agents. See local code graph for AI coding agents.
- Evidence memory (formerly MiseLedger) is Brigade’s built-in local evidence ledger over sessions and receipts, with crawl planning and receipt export. See local evidence ledger for agent work.
- Agent Pantry, Agent Notify, Skillet, and Memory Doctor cover auth sync, notifications, skills, and memory maintenance. Brigade built-ins and sidecars stay on brigade.tools; standalone kitchen tools use *.escoffierlabs.dev.
If you want to manage dotfiles in general
- chezmoi and GNU Stow are the standard answers. They are not agent-aware, but if you mainly want your shell and editor configs versioned, they are the right tool, not an agent layer.
If you want to run one task across several agents
- Brigade run takes one plain-language task, derives what it needs with a deterministic router (
brigade route "<task>"prints the plan first), and executes the required stages as dependency-ordered parallel waves across the CLIs you already pay for, then synthesizes the results into a receipt. The plan comes from task signals you can inspect and override (--route-signal +auth-surface), not from a model inventing its own steps.
How to choose
| Your job | Reach for |
|---|---|
| Same MCP servers in every coding agent | Brigade, or add-mcp for a quick reviewless copy |
| Memory inside a product you are building | mem0 or Letta |
| Memory across the coding-agent CLIs you run | Brigade or agentmemory (handoffs) |
| One reviewed source for MCP + tools + memory, with receipts | Brigade |
| Local code graph / impact for agents | Brigade code intelligence (formerly GraphTrail) |
| Local searchable agent work evidence | Brigade evidence memory (formerly MiseLedger) |
| Run one task across several agent CLIs | Brigade (brigade route derives signals, then runs dependency-ordered waves) |
| Just version my dotfiles | chezmoi or Stow |
The honest summary: if your problem is only one of these, a focused tool is lighter. Brigade is for the case where you have several coding agents and want their MCP servers, tools, and memory to come from one reviewed source, with a diff before every write and a receipt after.
Questions
Is Brigade the best memory tool? No, and it does not try to be. mem0 and Letta are stronger if your problem is programmable memory inside an app. Brigade is the best fit when you run multiple coding agents and want their config, tools, and memory unified and reviewable on disk.
What is the lightest way to just share MCP servers? add-mcp for a quick reviewless copy, or Brigade if you want the diff-and-receipt safety. See syncing MCP servers across coding agents, or the narrower Claude Code and Codex page if that is your exact setup.