Koltrix
AI & email
Reviewing and revoking AI app connections to your mailbox
A runbook for the life of an AI app connected to your email: how OAuth grants work, a quarterly review, when to revoke, what revoking does, admin off switches.
5 min read
Blog · AI & email
AI & email
AI triage, summaries, drafting and assistants that work with your mail, and how to keep a human in charge of every send.
AI & emailPrompting an AI assistant to search your mail effectively
How AI assistants turn your requests into mailbox searches, and how to phrase prompts with names, dates and domains to find the right thread the first time.
4 min read
AI & emailLeast privilege for AI email agents in a team workspace
An admin playbook for AI email agents: inherit user mailbox access, keep a workspace off switch, cut access on removal, scope by task, wall off sensitive mail.
4 min read
AI & emailRead, organize, draft, never send: a capability model for AI email
A five-tier capability model for AI in email: read, organize, draft, send and delete, with the risk and reversibility of each and why the line sits before send.
5 min read
AI & emailConnecting an AI assistant to your inbox: a pre-flight checklist
Before you connect Claude, ChatGPT or another AI assistant to your email, run this checklist: account, permissions, sending, revocation, admin rules and logs.
5 min read
AI & emailWhat is MCP? A plain-English guide for email users
MCP, the Model Context Protocol, explained for people who live in email: clients and servers, how connections work, and what an email server should expose.
5 min read
- AI & email
Defending AI email assistants against hidden instructions
Layered defenses against prompt injection in email: capability limits, human approval, untrusted-content handling, hidden HTML, tool chains and logging.
5 min read
AI & emailClassifying inbound mail by intent: a starter taxonomy
A ready-to-use set of intent categories for a SaaS inbox, with definitions, example emails and routing for each, plus rules for keeping the list short.
4 min read
AI & emailWho owns an AI-drafted reply in a shared inbox?
Answers to the accountability questions teams hit with AI drafts: who owns a reply, how to treat drafts, editing conflicts and whether to note AI use.
5 min read
AI & emailHow AI reads long email threads, and where it slips
Quoted history, signatures, context limits and who-said-what confusion: how AI models process long email threads, why summaries go wrong, and how to help them.
5 min read
AI & emailDrafting follow-ups with AI without sounding robotic
Before-and-after examples of AI-drafted follow-ups after silence, after a meeting and after a fix, with the edits that remove filler and add a clear next step.
5 min read
- AI & email
The limits of AI email sorting: edge cases to expect
Six situations where AI email sorting predictably gets it wrong, why each one fools the model, and the small habits and rules that cover the gaps.
5 min read