Stop holding it all
in your head
Local Mind reads your conversations with ChatGPT and Claude Code and keeps what you decided, still want and only assume — each with where you said it — held against your product. Your agents read it before they change anything.
ChatGPT export · Claude Code sessions · Held against your product · Read by your agents over MCP
10,000 credits to start · 1 credit per 10 lines of conversation
Your conversations with AI,
and the product they are about
Two sources today, each with one action — and the model of your product to hold them against.
Your export, once
Settings → Data controls → Export data, and drop the ZIP. Only conversations*.json is read, in your browser; images and files stay on your computer.
Every session, by itself
Drag your project’s folder once — or run one command, and each new session lands in Local Mind as you work.
The model it is held against
The repository you choose in GitMir Intelligence: every decision is checked against what the product actually does.
Back in a minute,
not an hour
The screen answers your questions first: what changed, what you decided, what waits for you.
What changed in your product, which of your decisions it touches, where the product now goes against one, and what waits for you — first on the screen.
A map of your decisions by area of your product. Open an area, then one decision as a chain: what it serves, the part of the product, what changed since your last visit, what the product does now. Every card says why it is there.
A timeline of what you decided, dropped and how the product’s answer changed — and your last visit beside today, side by side.
In force, cancelled, assumptions and what an assistant proposed — each with the conversation, the date and your own words.
In the product, the product does otherwise, not in the product yet, or not clear from the model — checked again after each new reading.
Over MCP your agents read the decisions in force about a part before they change it, and are told what you no longer want.
GitMir proposes three to seven goals from what you said; you confirm them, and every decision and agent action is shown under the goal it serves.
Each session and push of an agent, held against what you meant: matched it, went against a decision, or changed parts nobody named.
Once a week: your decisions and what changed in the product after each — only when there is something to say.
The same sheet for a lead, a product person or an engineer — the same parts, the words of your job.
Your Local Mind is yours. Share one decision or all in force, and your team and their agents see it with your name.
The text of your conversations,
nothing else
Only what you and the assistant said, and “[edited <path>]” when an agent changed a file. Images, files, audio, tool output and subagent sessions stay on your computer. Keys, tokens and passwords are removed before upload, and again when GitMir receives the text. It is processed in GitMir’s cloud; delete any conversation, or everything, at any time.
Paid in credits
10 lines of conversation analysed are 1 credit, counted over everything you add; a conversation you already added only adds its new lines. An agent’s question is 1 credit, like any MCP question. 10,000 credits to start, shared with your team’s other GitMir products.
Your thinking,
kept straight
The conversation, the date and your own words — one click away.
When something you said disagrees with what you decided earlier, both sides are put in front of you. You decide which one stands.
A later “no longer” cancels the earlier decision, and your agents are told not to bring it back.
Claude Code sessions arrive by themselves after one command; a new ChatGPT export adds only what is new.
Before you bring in your conversations
What is GitMir Local Mind?
Local Mind reads your conversations with ChatGPT and Claude Code and keeps what you decided, still want and only assume — each with where you said it — held against your product. Your agents read it before they change anything.
Which conversations can it read?
The ChatGPT data export — Settings → Data controls → Export data, the ZIP that comes by mail — and Claude Code sessions, the folder of your project in ~/.claude/projects. From Claude Code, one command makes every new session arrive by itself.
What leaves my computer?
Only the text of your conversations: what you and the assistant said, and “[edited <path>]” when an agent changed a file. Images, files, audio, tool output and subagent sessions stay on your computer. Keys, tokens and passwords are removed before upload, and again when GitMir receives the text. It is processed in GitMir’s cloud.
How does it know what I decided?
Only what you said or agreed to counts as your decision; what an assistant proposed waits for your Confirm. When you later say “no longer” or “instead”, the earlier decision is marked cancelled, with the line that cancelled it. When two things disagree without that, Local Mind asks you which one holds — it never chooses for you.
How is it held against my product?
Each decision in force is checked against the model of the repository you choose in GitMir Intelligence — in the product, the product does otherwise, not in the product yet, or not clear from the model — and again after each new reading. “Since your last visit” says what changed and which of your decisions it touches.
How do I see how my decisions connect?
On the map. Your decisions are drawn by area of your product, always in the same place; open an area to see what you want, decided and assume there with the parts of the product they are about, then open a decision to see it as a chain — what it serves, the part, what changed since your last visit and what the product does now. A timeline shows week by week what you decided and dropped and how the product’s answer changed, and Then / now puts your last visit beside today. The next visit opens where you stopped.
Do my agents see it?
Over MCP. gitmir_decisions gives your agent the decisions in force about a part before it changes it, and what you no longer want; gitmir_decide writes down a decision you just told it, as a proposal you confirm. Your own keys see your decisions; teammates and their agents see only what you shared.
Can my team see it?
Not until you share — one decision, or all in force. A shared decision shows to everyone who reads that project, with your name, and you can stop sharing at any time.
How much does it cost?
10 lines of conversation analysed are 1 credit, counted over everything you add; a conversation you already added only adds its new lines. An agent’s question is 1 credit, like any MCP question. 10,000 credits to start; credits are shared with your team’s other GitMir products.
Can I delete what it keeps?
Any conversation, or everything, from the Local Mind screen. What came only from a deleted conversation goes with it.
What are goals in Local Mind?
GitMir proposes three to seven goals from what you said in your conversations. You confirm, rename or drop them, and your decisions and your agents’ actions are shown under the goal each one serves.
Does Local Mind send me email?
Once a week, and only when there is something: your decisions of the week and what changed in the product after each one.
Bring in your conversations
Keep your decisions
Drop the ChatGPT export or a Claude Code folder, and see what you decided — held against your product.