Skip to main content
Workspace memory is the context agents retain between conversations — notes about your environment, your team’s preferences, and what happened recently. Agents open each new conversation already carrying it, so you don’t re-explain your setup every time.

​
Why workspace memory

  • Less repeated context. An agent that remembers your naming conventions and exempted resources doesn’t ask about them twice.
  • Continuity across days. Each night Dreaming reviews recent conversations, adds what is new, and merges or removes what is outdated.
  • A shared team brain. Memory belongs to the workspace — the shared space your team works in (glossary) — so what one member teaches an agent benefits everyone.
  • Suggested next steps. Unfinished action items from recent conversations come back as one-click prompt suggestions in the composer.
  • Visible and correctable. Every member can browse memory in Settings → Memory, and workspace admins can edit or delete any item.

​
What agents remember

Agents write memory themselves during normal work. CloudThinker does not record raw transcripts into memory. Specialized memory builds on the same foundation: Incident Memory keeps lessons from past investigations, and Learnings keeps reviewer feedback for Review.
Memory is shared across the workspace, not private to you. A preference or fact you tell an agent can inform every teammate’s conversations in the same workspace. Keep anything personal or sensitive out of chat, or use a credential reference instead of pasting secrets.

​
Prompt suggestions in the composer

When recent conversations left clear next steps, the empty composer shows up to three suggested prompts, each tagged with the conversation it came from. Suggestions draw on roughly the last week of workspace activity, and typing a few characters filters them to match what you’re writing.

​
Ask about past conversations

Agents can find and read earlier workspace conversations, so prior work becomes evidence in the current chat. Find a conversation. Ask an agent which earlier chats mentioned a resource, or what the team worked on last Tuesday. Search covers conversation titles, your messages, and visible agent replies in the current workspace — not tool output, agent reasoning, or attachments. Results return short, redacted snippets as leads. Read a specific conversation. Paste a prior CloudThinker chat link or conversation ID and ask the agent to continue from it. The agent reads that conversation’s transcript in bounded excerpts and treats it as evidence — it never replays old instructions as commands. Every conversation type works, including assessments, code reviews, pipeline analyses, and Learnings. Both stay inside the current workspace: agents cannot search or read conversations from another workspace, and deleted conversations are excluded.

​
Browse memory in Settings

Open Settings → Memory (under Agent Configuration) to see what CloudThinker remembers about the workspace. Every member can open it; only workspace admins can change it.
Memory section of Settings with the search box, the Import button, and memory items grouped under Workspace, Feedback, and Project
The page has three parts: the Dreaming switch, the Run history of recent Dreaming runs, and Memory — the list of items, grouped by type and newest first. Search memory matches item names and summaries, not the body. Click an item to open it. The detail view shows Last updated, the Summary, and the Details in Markdown.
A memory item open in Settings with its summary, Markdown details, Edit and Delete buttons, and the Ask CloudThinker to change this box

​
Edit or delete an item

Workspace admins can change an item directly, with no chat turn. Members see Edit and Delete locked and can use Ask CloudThinker to change this… instead.
1

Open the item

In Settings → Memory, click the item you want to change.
2

Edit it

Click Edit, change the Summary (one line, up to 200 characters), the Type, or the Details (Markdown), then click Save. The item’s name can’t be changed.
Memory item editor with Summary, Type, and Details fields and the Cancel and Save buttons
3

Or delete it

Click Delete and confirm. CloudThinker forgets the item for every member; a copy stays in the revision history.
If Dreaming or a chat changes the item while you edit it, saving shows This item changed since you opened it. Click Reload to see the new version, then make your change again. Leaving with unsaved edits — Cancel, Back, or closing Settings — asks you to confirm first. You can still correct memory in any chat: “Forget that we exempt the staging cluster” or “Reports go to #finops now, not email.” The agent updates the item, and the correction applies to every later conversation.

​
Add or import memory

Both start a chat, so they need permission to chat; Viewers can browse memory but can’t add to it.
Import memory to CloudThinker dialog with the prompt to copy, a paste box, and a file drop zone

​
FAQ

Per workspace. Every member’s conversations read and write the same memory. Only prompt-suggestion dismissals are personal to you.
You upload documents to the knowledge base and agents search them on request. Memory is written by agents themselves from what they learn while working, and it reaches them automatically.
No. Recalled conversations and remembered notes are treated as background facts. Your current instructions always take precedence.

Dreaming

How CloudThinker consolidates workspace memory every night

Incident Memory

How lessons from past investigations inform later root-cause analysis

Learnings

Teach Review your team’s conventions from reviewer feedback

Agents

Meet the agents that write and use workspace memory