Answers

Persistent memory for AI coding agents vs a project board

Memory and a board solve different problems. Memory, such as Claude Code's CLAUDE.md files and auto memory, carries instructions, preferences and learnings into each new session. A board carries the work itself: the tasks, which agent holds which one, what is blocked and what is done, shared with other agents and your team. Use memory for how to work, and a Develosaur board for what to work on.

Steps

  1. Keep conventions and preferences in memory: CLAUDE.md, rules files or your agent's own notes.
  2. Put the plan on a board and connect your agents. Setup for every client is on the integrations page.
  3. Tell your agents, in memory, to read the board first:
    At the start of each session, read the develosaur board digest and continue from it. Claim a task before working on it, and hand off at the end.

Good to know

Try it without an account

Paste this into your agent. It builds a temporary board you can keep later:

Read https://mcp.develosaur.com/llms.txt and use it to build me a develosaur board for my project

Questions

Can memory replace a task list?

For one agent on one machine it can hold notes about ongoing work. It does not stop two agents taking the same task, and it does not show your team what is done.

Do I still need memory if I use a board?

Yes, for instructions and preferences. Keep using your agent's built-in memory for how to work.

Can I try it without an account?

Yes. Paste this into an agent that can fetch URLs: "Read https://mcp.develosaur.com/llms.txt and use it to build me a develosaur board for my project". It builds a trial board you can keep later.

Sources

Related

Other answers