Space Memory

Keep the decisions, risks, conventions, and context behind your work. People and AI agents add entries, link them to tasks and code, and find them later with natural-language search.

Try free for 14 daysBook a demoStart with Google

Space Memory

Keep decisions, risks, and conventions from your team and AI agents linked to tasks and code.

Key capabilities

Your space never forgets

Typed entries

Every entry says what kind of knowledge it holds, so the list is easy to scan.

  • Six kinds: Decision, Context, Convention, Problem, Risk, and Session log
  • Lifecycle: Mark entries as Active, Resolved, Obsolete, or Superseded
  • History: See how each entry changed over time

Linked to tasks and code

Keep each entry next to the work it explains.

  • Tasks: Open the Memory tab of a task to see the entries linked to it
  • Code: Link pull requests, merge requests, branches, and commits
  • Projects: Review the memory of a whole project in one view

Written by people and agents

Your team and AI agents add to the same memory.

  • Clear source: Every entry shows who added it, including the AI tool an agent used
  • Agent access: AI agents read and write memory through MCP tools with dedicated permissions
  • Coding agents: Relevant team memory is included when you delegate an issue

Natural-language search

Find past decisions and known risks by describing what you need instead of guessing the exact title.

Context that outlives the conversation

Stop explaining the same decisions twice

Decisions often live in chat threads, meeting notes, and the heads of the people who made them. Space Memory gives them a permanent place next to the work. When someone joins the project or an agent picks up a task, the reasons behind earlier choices are already there.

Decisions: Record what the team chose and why. Risks and problems: Keep known issues visible until they are resolved. Conventions: Write down the rules people and agents should follow.

Memory for AI agents

Agents that remember your project

AI tools connected through the Vaiz MCP server can read existing entries before they start and save what they learn as new ones. When you delegate an issue to a coding agent, Vaiz adds relevant team memory to its context, so the agent follows your conventions and knows about existing risks.

Common questions

Decisions, context, conventions, problems, risks, and session logs. Each entry can be linked to tasks and code.

Yes. Agents connected through the Vaiz MCP server read and write entries with dedicated permissions, and each entry shows which AI tool added it.

Mark the old entry as Superseded or Obsolete. Its history stays available, so you can see what changed and when.

Search in plain language. Vaiz finds entries by meaning, not only by matching words.

Space Memory is a power-up for your space. To use it from AI tools, connect Vaiz through MCP.

See it in action in your own workspace.

Start a free trial, set up your first project in minutes, and invite your team when you’re ready.

Try free for 14 daysBook a demo