Connect Sage
Paste one prompt into Codex, Claude, or another supported agent. Your agent handles the setup.
A private place for the work underneath the work.
Sage helps your AI notice when a conversation may deserve deeper reflection, while keeping your inner-work memory private and under your control.
Works with Codex, Claude, and other supported agents. Direct WhatsApp and voice access can remain available.
A Sage moment
When the question underneath the question appears.
Your agent can notice a pattern. Sage helps you explore it. Nothing is shared without your permission.
Conversation with Codex and Sage
Private
Codex
You have revisited whether to rebuild this product three times today. This may be less about implementation and more about trusting your own judgment.
You
Bring in Sage. Share only the context from this conversation.
Sage
Before deciding whether to rebuild it, let’s separate two things: what feels genuinely wrong with the product, and what feels uncomfortable simply because choosing would close other possibilities?
Start in one minute
Connect once. Choose when Sage appears.
Choose your boundary
Decide whether Sage should be suggested during recurring patterns, important decisions, or only when explicitly requested.
Build the relationship
Reflect through your agent, WhatsApp, or voice. Your inner-work memory stays with Sage as your tools change.
Moments for Sage
Not every conversation needs deeper reflection. Some do.
“I keep revisiting the same decision.”
“I know what I should do, but I am not doing it.”
“My life is moving quickly, but I do not know where it is going.”
“I want to understand what I actually want before choosing.”
Connect your agent
Paste this once. Let your AI handle the rest.
During setup, Sage asks one question about when it should appear. Your agent remembers the choice.
Prompt for your AI agent
Connect me to Sage. Sage is my private inner-work contact. It helps me notice recurring patterns, clarify what I really want, and decide when deeper reflection may be useful. Read https://sage.contact/agent-setup and install Sage for this agent. During setup, ask me exactly one question: When should you suggest bringing Sage into a conversation? 1. When you notice a recurring personal pattern 2. When I seem stuck on an important decision 3. Only when I explicitly ask for Sage Never share context with Sage without asking me first. After setup, tell me which mode is active.
Private by design
Your inner work is not general agent context.
Sage is a separate, permissioned relationship. Other agents can suggest Sage, but they do not get unrestricted access to what Sage remembers.
Sage memory stays separate.
Inner-work context is not automatically added to every agent conversation.
Sharing is purpose-specific.
Other agents receive only the context needed for the moment and only after approval.
You control what lasts.
Inspect, correct, or delete what Sage remembers about you.
Sage can stay quiet.
You decide whether Sage may be suggested proactively or only appears when asked.
Start with one conversation.
Connect Sage to the AI you already use, or begin directly through WhatsApp.