--- title: "Test your Microsoft 365 and GitHub Copilot context." description: "Test Microsoft 365 Memory separately from GitHub repository instructions. Keep project decisions in maintained files each product can read." canonical: "https://scalewithsearch.com/articles/why-does-copilot-forget" date: "2026-01-28" modified: "2026-10-09" --- ## Site navigation - [Scale With Search](https://scalewithsearch.com/) - Real estate - Real estate - [Real estate](https://scalewithsearch.com/for/real-estate) - Work - Start here - [Send your brief](https://scalewithsearch.com/work#send-your-brief) - [Prepare your six-question brief](https://scalewithsearch.com/work#prepare-your-six-question-brief) - Build - [Site build, content library with SEO, signal desk](https://scalewithsearch.com/work) - For your business - Trades and home services - [Auto body and collision shops](https://scalewithsearch.com/for/auto-body-and-collision-shops) - [Foundation and home repair contractors](https://scalewithsearch.com/for/foundation-and-home-repair) - [Garage door and fencing contractors](https://scalewithsearch.com/for/garage-door-and-fencing-contractors) - [HVAC contractors](https://scalewithsearch.com/for/hvac-contractors) - [Janitorial and commercial cleaning companies](https://scalewithsearch.com/for/janitorial-and-commercial-cleaning) - [Locksmiths](https://scalewithsearch.com/for/locksmiths) - [Moving companies](https://scalewithsearch.com/for/moving-companies) - [Pest control companies](https://scalewithsearch.com/for/pest-control-companies) - [Plumbing and electrical contractors](https://scalewithsearch.com/for/plumbing-and-electrical-contractors) - [Restoration and water or fire damage companies](https://scalewithsearch.com/for/restoration-and-water-fire-damage) - [Roofing companies](https://scalewithsearch.com/for/roofing-companies) - [Towing companies](https://scalewithsearch.com/for/towing-companies) - [Tree services and landscaping companies](https://scalewithsearch.com/for/tree-services-and-landscaping) - [Solar installers](https://scalewithsearch.com/for/solar-installation) - [General contractors](https://scalewithsearch.com/for/general-contractors-and-construction) - [Paving, concrete, and flooring contractors](https://scalewithsearch.com/for/paving) - Practices and professional services - [Bookkeeping and tax practices](https://scalewithsearch.com/for/bookkeeping-and-tax-practices) - [Dental practices](https://scalewithsearch.com/for/dental-practices) - [Family and criminal defense law firms](https://scalewithsearch.com/for/family-and-criminal-defense-law-firms) - [Med spas and aesthetics practices](https://scalewithsearch.com/for/med-spas-and-aesthetics) - [Personal injury law firms](https://scalewithsearch.com/for/personal-injury-law-firms) - [Veterinary clinics](https://scalewithsearch.com/for/veterinary-clinics) - [Gyms and fitness studios](https://scalewithsearch.com/for/fitness) - [Therapy and outpatient health practices](https://scalewithsearch.com/for/therapy-and-outpatient-health) - [Medical billing companies](https://scalewithsearch.com/for/medical-billing) - [Insurance agencies](https://scalewithsearch.com/for/insurance-agencies) - [Financial advisors](https://scalewithsearch.com/for/financial-advisors) - [Property management companies](https://scalewithsearch.com/for/property-management) - [Recruiting and staffing agencies](https://scalewithsearch.com/for/recruiting-and-staffing) - [Architects and interior designers](https://scalewithsearch.com/for/architects-and-interior-designers) - [Logistics and supply chain companies](https://scalewithsearch.com/for/logistics-and-supply-chain) - Agencies, MSPs, and manufacturing - [IT and managed service providers](https://scalewithsearch.com/for/it-and-managed-service-providers) - [Machine shops and precision manufacturers](https://scalewithsearch.com/for/machine-shops-and-precision-manufacturing) - [Marketing agencies and freelancers](https://scalewithsearch.com/for/marketing-agencies-and-freelancers) - [SEO agencies and consultants](https://scalewithsearch.com/for/seo-agencies-and-consultants) - [Small manufacturers and fabricators](https://scalewithsearch.com/for/small-manufacturers-and-fabricators) - Restaurants, shops, studios, and nonprofits - [Restaurants and hospitality businesses](https://scalewithsearch.com/for/restaurants-and-hospitality) - [Retail stores and ecommerce sellers](https://scalewithsearch.com/for/retail-and-ecommerce) - [Photographers, event planners, and travel agents](https://scalewithsearch.com/for/photographers) - [Churches and nonprofits](https://scalewithsearch.com/for/churches-and-nonprofits) - [All industries](https://scalewithsearch.com/for/) - Learn - For your office - [Office job guides](https://scalewithsearch.com/guides/) - [Browser calculators](https://scalewithsearch.com/tools/) - Start here - [How it works](https://scalewithsearch.com/how-it-works) - [Free Starter Kit](https://scalewithsearch.com/kit/business-memory-starter-kit.zip) - [Synthetic specimen](https://scalewithsearch.com/specimen/working-session-specimen.zip) - Guides - [The Complete Guide to Business Memory for AI Agents](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [The Complete Small-Business Guide to AI Agent Governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [The Complete Guide to Leaving Vendor AI Memory](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - Articles by cluster - [Business memory](https://scalewithsearch.com/articles/business-memory-for-ai-agents-guide) - [Agent governance](https://scalewithsearch.com/articles/ai-agent-governance-guide-small-business) - [Migration and ownership](https://scalewithsearch.com/articles/leaving-vendor-ai-memory-guide) - For machines - [llms.txt](https://scalewithsearch.com/llms.txt) - [llms-full.txt](https://scalewithsearch.com/llms-full.txt) - [Machine view](https://scalewithsearch.com/?view=machine) - Company - Evidence - [Proof](https://scalewithsearch.com/proof) - Company - [About](https://scalewithsearch.com/about) # Route Microsoft 365 and GitHub Copilot context to the right memory layer. You have worked with GitHub Copilot on a project for weeks. You explained the team structure, the naming rules, and the code standards. Today you open a new GitHub Copilot chat, and it asks which language you use. That code chat belongs to GitHub Copilot. Microsoft 365 Copilot is a separate product for Word, Excel, Teams, Outlook, and related work. Microsoft 365 Memory does not supply GitHub repository instructions. This guide routes each task to the product and context layer it needs. For Microsoft 365 recall diagnosis alone, use [the Microsoft Copilot memory test](/articles/copilot-memory-problems). ## Separate Microsoft 365 file access from its memory Microsoft 365 Copilot connects to Word, Excel, Teams, Outlook, and the rest of your Microsoft 365 environment. It can pull data from documents, summarize email, and search SharePoint. That is access to files as they are now. It is not memory of what you discussed. Ask Microsoft 365 Copilot to analyze a spreadsheet. Close the chat. Open a new one and ask about the same spreadsheet. Access to the workbook does not establish that the new chat has loaded the first analysis or your conclusions. Keep those conclusions in a maintained document when the next task needs them. Each conversation starts independent, by design. A memory feature adds a selected layer on top. ## Know what Microsoft 365 Copilot Memory keeps In July 2025, Microsoft announced Copilot Memory for Microsoft 365 Copilot ([Microsoft Tech Community](https://techcommunity.microsoft.com/blog/microsoft365copilotblog/introducing-copilot-memory-a-more-productive-and-personalized-ai-for-the-way-you/4432059)). The announcement described four behaviors: - Memory is on by default for users with a Microsoft 365 Copilot license. - Microsoft 365 Copilot picks up details from conversations, such as a preferred language or a current project. - It stores those details in a hidden folder in the user's Exchange mailbox. - It shows a "memory updated" signal when it saves something, and you manage saved memories in settings. That helps with small, repeated preferences. Three limits remain: - **It saves discrete facts.** Microsoft 365 Memory keeps preferences and project names. It does not keep the reasoning, or the three approaches you tried last month and why you rejected them. - **It does not replay conversations.** Microsoft 365 Copilot keeps a chat history you can scroll, but a new chat does not load it. You find the old chat and reference it yourself. - **It keeps methods only as separate rules.** Microsoft 365 Memory may save a stated style preference. It does not keep how you structure projects, name variables, or handle edge cases. Each of those needs its own saved memory. The [ChatGPT and Microsoft 365 Copilot memory comparison](/articles/chatgpt-vs-copilot-memory-microsoft-365) tests both products against the same business record. ## Use GitHub Copilot repository instructions for code work GitHub Copilot code work has a repository option: an instruction file that lives with the project. It is separate from Microsoft 365 Memory in Exchange. GitHub Copilot reads a file named `.github/copilot-instructions.md` from the repository. GitHub Copilot Chat in VS Code, Visual Studio, and on GitHub uses it as standing instructions for that repository. The team checks the file into version control, so every developer gets the same context, and each change has a history. In Visual Studio, the GitHub Copilot Memories feature, described on the Visual Studio blog in January 2026, goes a step further. When you state a preference in chat, it can offer to save it to your personal instructions or to the repository instructions file ([Visual Studio blog](https://devblogs.microsoft.com/visualstudio/copilot-memories/)). That is close to real memory: a readable file, in the repository, under version control. It covers code in a repository. It does not cover the Microsoft 365 Copilot chat in Word or Outlook. In those Microsoft 365 apps, use their session context, referenced documents, and Microsoft 365 Memory instead. A short repository file can look like this: ```markdown # Project conventions - Language: TypeScript, strict mode. - Names: camelCase for functions, PascalCase for types. - Tests: every new function gets a unit test in /tests. - Do not add new dependencies without a note in DECISIONS.md. - Rejected: an ORM for the reporting queries (too slow at our row counts). ``` The last line matters. A rejected option is exactly the context that chat memory drops. ## Diagnose Microsoft 365 context workarounds separately For Microsoft 365 Copilot, people try four fixes before they maintain context in files: - **Manual memory saves.** You tell Microsoft 365 Copilot "remember this" after each important detail. This turns into memory management. You curate what the AI knows. - **An uploaded context document.** You keep a "Microsoft 365 context" document and add it to each chat. It works until the document is long enough that Copilot uses parts of it less. - **SharePoint as a knowledge base.** Microsoft 365 Copilot can search SharePoint. It does not know which documents matter for this task unless you say so. - **History review.** You scroll old chats and copy exchanges into a new one. It is slow, and a long paste crowds the conversation. Each fix breaks for the same reasons. Microsoft 365 Memory may keep a choice and lose the reason, and then suggest the rejected options again. Memories do not compose. Microsoft 365 Copilot can know that you use Python, work on Project Alpha, and prefer asynchronous code. It can still fail to combine them into one request. And you spend time on memory upkeep, which is the time the tool was meant to save. ## Keep owned files for Microsoft 365 and GitHub tasks For Microsoft 365 work outside a repository, keep one maintained context file. For GitHub code work, keep the decisions file beside the repository instructions. Record who you are, what you build, how you work, what you tried, and what you rejected. Store it where you work, and update it when the project changes. Claude Code works this way outside the Microsoft stack. It is a terminal program that reads a `CLAUDE.md` file from the project folder at the start of every session. The same file works for a team when it is under version control. Nobody needs to ask whether someone told the AI about the naming rule. The [guide to Claude Code](/articles/what-is-claude-code) explains that product choice. Inside Microsoft 365, the chat apps do not read a local folder when a chat starts. You reference the file in the prompt, keep it in OneDrive or SharePoint where Microsoft 365 Copilot can find it, or paste it. The file still gives you one maintained source. Microsoft 365 Memory then carries selected preferences, and the file carries the facts and decisions. Microsoft 365 Copilot fits teams whose documents live there. GitHub Copilot fits repository work. If working context lives in local files or other tools, give the chosen product an explicit path to a maintained source. Neither product name alone proves the needed file loaded. ## Match Microsoft 365 and GitHub context to a layer Each layer holds one kind of context well. Match the context to the layer before you blame the tool. | Context | Layer that holds it | Why | |---|---|---| | Microsoft 365 personal preferences, such as summary length or tone | Microsoft 365 Copilot Memory | Selected facts you manage in Microsoft 365 settings | | GitHub code conventions for one repository | `.github/copilot-instructions.md` | Shared, versioned, and read by GitHub Copilot Chat | | Microsoft 365 or GitHub project decisions and rejected options | A decisions file accessible to that product | Keeps the reason explicit instead of relying on selected memory | | Microsoft 365 client facts, prices, and approvals | A maintained current-file record with an owner | These need a source and a review date | | Microsoft 365 or GitHub working state of a long task | A checkpoint file accessible to that product | A new chat needs the current state supplied | If a fact moves between layers, write it in the file first. Then update or delete any corresponding Microsoft 365 Memory entry. For GitHub, update the repository instructions when the standing convention changes. Keep the sources from disagreeing. ## Test Microsoft 365 Memory and GitHub instructions separately Run steps 1 through 6 in Microsoft 365 Copilot. Run the repository check in GitHub Copilot separately. The two tests cannot establish memory sharing between products. 1. In a new Microsoft 365 Copilot chat, state a harmless preference, such as "Summaries use three bullets." 2. Confirm that Microsoft 365 Copilot shows the "memory updated" signal. 3. In the same chat, discuss two options and reject one, with a reason. 4. Open a new Microsoft 365 Copilot chat and ask for a summary. Check the bullet count. 5. Ask which option you rejected and why. 6. Open Microsoft 365 Memory settings and read what it saved. 7. For code, add one rule to `.github/copilot-instructions.md`. Open GitHub Copilot Chat in that repository, ask a question it affects, and check the answer. Record whether each step passes. If Microsoft 365 loses the rejected option or reason, keep it in the maintained document. If GitHub misses a repository rule, check the active repository, instruction support, and model adherence before changing the rule. The [guide to types of memory in AI systems](/articles/types-of-memory-ai-systems-business) maps each result to the layer that should hold it. The same owned-record method runs each client's content library with SEO and signal desk: [How the build works: Owned files, checks, handoff.](/how-it-works) ## Related: AI memory - [Test what Microsoft Copilot remembers and keep the rest in a file you own](/articles/copilot-memory-problems) - [Make Microsoft Copilot remember your work with its memory settings and a context file](/articles/how-to-make-copilot-remember) - [ChatGPT vs Microsoft Copilot Memory: Which One Should Prepare the Next Microsoft 365 Job?](/articles/chatgpt-vs-copilot-memory-microsoft-365) ---- ```text .|########||. .|########||. .|########||. |##||. .||##|. |##||. .||##|. |##||. .||##|. |#|. .|#|. |#|. .|#|. |#|. .|#|. |#| |#| |#| |#| |#| |#| .#| |#. .#| |#. .#| |#. .#. .#| .#. .#| .#. .#| |#. .#| |#. .#| |#. .#| |# ...... #| |# ...... #| |# ...... #| .# ||#########| #| .# ||#########| #| .# ||#########| #| .#. |######||######|. .#. .#. |######||######|. .#. .#. |######||######|. .#. #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# #. .##|###|##|#|######| .# || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || || |##|#||||||||||||#||#| || # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. # |#||||||||||||||||||||#| #. || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # # .#||||||||||||||||||||||||#| # || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || || ....|||#||||||##|#|||#|#||##||||....|. || #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# #. . ....|# ....#||| ||| .#||| ....#. .# # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. # . ||||#| .#####|| . .#| .#||| ||||# #. .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || .| ||||| #. |#|||||. || #. |#||. .||||| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || || |.... #. ....|#. |. ...|. ....|| || #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# #. .||||||##||||||##||####|||||||#||||||#| .# # .||####|###########################|. # # .||####|###########################|. # # .||####|###########################|. # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#||||||.#.|| # |. ..# #| #|||||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||||| .. |# ## |#| ...#|#|||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#|||#|# .# .#| #| ##|.#..#|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # .#||||||############|######|||||#| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # # |...||#||||||#||||#|#||||||#|||||#||| |.| # #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# #. .. ||||# .|||##|. |#| .|| || .|||#. #| # .# || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || | ...|| ...##| | #| .|. |. ##### .. .| || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || || ||||| || ||||#| . |. |. |#. ||||| |#| || || .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. .# |.....#|....|#.||||.|||##.|#|....||.#||.#. #. #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# #.|||||||#############################| ||| .# || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || || ##||#||#||||||#||#||#||#||##. || # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # # .#|||||||||#||#|||||||||||#| # || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || || |#||||||||||||||||||||||#| || .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. .# |#||||||||||||||||||||#| #. || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || || |#||||||||||||||||||#| || #. .######|#|#########| .# #. .######|#|#########| .# #. .######|#|#########| .# .#. |#####||||#####| .#. .#. |#####||||#####| .#. .#. |#####||||#####| .#. |# ||########|| #. |# ||########|| #. |# ||########|| #. |# ...... #| |# ...... #| |# ...... #| |#. .#| |#. .#| |#. .#| |#. .#. |#. .#. |#. .#. .#| |#. .#| |#. .#| |#. |#| |#| |#| |#| |#| |#| .|#|. .|#|. .|#|. .|#|. .|#|. .|#|. |##||. .||##| |##||. .||##| |##||. .||##| .||########||. .||########||. .||########||. Scale With Search 2026 [scalewithsearch.com](https://scalewithsearch.com) ```