Memory
How a Track keeps its decisions, commitments and know-how, so the next person picks up where things left off.
Memory keeps continuity around long-running work.
Every Track has its own memory: the decisions made, who committed to what, the ideas on the table, open questions, preferences, facts and procedures for that piece of work. When you or Giga start work in the Track, Giga reads its memory first, so nobody has to re-explain where things stand.
For the full picture of how memory works across Giga, see Giga Memory.
How a Track's context is structured
| Layer | Purpose |
|---|---|
| Context Notes | Durable background and standing instructions |
| Memory | Decisions, commitments, ideas, requests, preferences, facts and procedures, kept current |
| Evidence | The append-only record behind each memory: what was said, by whom, when, and where |
| Files | Authored work product such as reports, drafts, code and data |
Memory stays current
Each memory is one clear statement with a type, such as a decision or a commitment. As new information arrives, Giga updates the Track's memory rather than piling up notes:
- a new decision replaces the old one, and the old one is kept as history
- a commitment is closed when someone says it's done
- the same point heard again adds evidence instead of a duplicate
So "where are we on this?" has a current answer: the Track's open commitments, latest decisions and open questions.
Evidence
Behind every memory is its evidence: append-only records of what was actually said, with a quote, who said it, when, and where it came from (a chat, a saved passage, pasted notes or an imported meeting). Evidence is never rewritten, so you can always see how the Track's memory got to where it is.
How memory gets into a Track
- From chats in the Track. When you work with Giga inside a Track, lasting points from the conversation are saved to that Track.
- When you ask. In ChatGPT, Claude or Slack, say "save this" and choose the Track.
- By filing. Anything saved without a Track waits, private to you, in Brain → To file until you file it to a Track.
- From notes and meetings. In Brain → Add to memory, paste notes or import a meeting link and choose the Track.
Why this matters
A useful memory helps Giga answer:
- What has already been decided?
- What changed since we last worked on this?
- Who is doing what, and by when?
- What's still open?
- What has already been tried?
Memory vs Context Notes
Context Notes and Memory work together but do different jobs.
| Use a Context Note when… | Use Memory when… |
|---|---|
| The information should stay true across future work | It's a decision, commitment, change or learned fact |
| It's a standing instruction or constraint | It came up during the work |
| It explains what the Track is about | It explains how the Track has evolved |
| You want to curate the background deliberately | You want continuity from the work itself |
A useful rule of thumb:
Context Notes explain the work. Memory explains what happened to the work.
Memory vs files
Files are authored work product. Memory is structured knowledge about the work. A Track can have rich memory even when its file folder is empty, and a folder full of files doesn't tell Giga which decision was final, what changed recently, or what still needs attention.
Memory and permissions
A Track's memory is only available to the people on that Track, in Giga Web, Slack, ChatGPT and Claude.
Giga never picks a Track for you. Saving to a Track shares it with everyone on that Track, so Giga asks you to choose, suggests the most likely Track, or keeps the memory private until you file it.
A simple example
Imagine a Track for Website redesign.
Context Note
The audience, positioning, brand constraints and standing design principles.
Decision
The team decided to remove the pricing calculator from the launch scope.
Commitment
Sam will have the pricing page ready for review by Friday.
Files
Copy drafts, screenshots, design exports and implementation files.