Open a private security advisory on this repository. Please do not open a public issue for a vulnerability.
You will get an acknowledgement within 72 hours and an assessment within seven days. There is no bounty programme — this is a single-maintainer project — but every report is credited in the advisory unless you ask me not to.
teams-draft-assistant drafts replies for your Microsoft Teams 1:1 chats. You review
and send. It does not auto-reply and it does not impersonate you.
That sentence is the product, and the threat model is built around defending it.
| Class | Why it matters |
|---|---|
| Anything that sends without you | A path that dispatches a message without an explicit human action is the most serious bug this project can have. The entire justification for this tool existing is that a person decided, every time. |
| Anything that impersonates you | Presenting generated text as authored by you, anywhere, without the review step. |
| Conversation content leaving the machine unexpectedly | Chat contents are other people's words, and they never consented to anything. Any transmission beyond what the operator explicitly configured is in scope. |
| Token or credential mishandling | An access token written to disk, printed, logged, or included in a crash report. |
| A scope request wider than the work | Asking for Graph permission the tool does not need — including any write scope beyond what sending a reviewed message requires. |
- Draft quality. That is a feature request, not a vulnerability.
- Requests for automatic replies. That is a deliberate design decision and it will not change; it is the difference between an assistant and an impersonator.