Overview
Most agents start from the task. Elephant Agent starts from the person.
Mother starts from the person, not the task. She grows a correctable Personal Model of who you are, what surrounds you, what is alive right now, and what your path has taught you. That understanding keeps deepening through interaction, correction, and gentle questions.
Once Elephant Agent understands enough, it can help design living Paths: work, health, habits, learning, relationships, recovery, research, code, and other long-running directions you want to move.
That is the fourth level Elephant Agent is aiming at. L1 agents execute tasks. L2 agents carry context. L3 agents improve procedures. Elephant Agent's L4 position is that personal AI should help the person grow over time, not just automate more sessions.
The mechanism is a correctable Personal Model. Memory is the beginning, not the product. The center is what Elephant Agent currently understands about your Identity, World, Pulse, and Journey, with source evidence and open questions close enough to inspect and correct. Paths are how that understanding becomes useful in daily life.
If you are on macOS, start with the desktop app. If you are on Linux, cloud, SSH,
or a terminal-first setup, install CLI + Dashboard, run elephant init, then
return through elephant wake.
Choose your path
| If you want to... | Read this first | What you will understand |
|---|---|---|
| Start using Elephant Agent | Quickstart | Choose macOS app or CLI + Dashboard. |
| Understand Paths | Paths, Steps, and Herds | How Mother turns understanding into Paths, Steps, Flow, Checkpoints, and Herds. |
| Use the primary product | macOS app | Chat / Wake, Paths, Personal Model, Herd, skills, messaging, calendar, usage, and settings in one workspace. |
| Use a terminal or remote machine | CLI / Chat TUI | How elephant, init, wake, slash commands, and dashboard --no-open fit together. |
| Inspect what it knows | Dashboard | How Personal, Agent, and System pages map to implementation surfaces. |
| Understand the thesis | Why Elephant Agent | Why Elephant Agent keeps judgment, evidence, questions, and learning close to the person. |
| Extend what it can do | Skills and Tools | How visible capabilities orbit the Personal Model. |
The core idea
| Product bet | What it means | Where to go deeper |
|---|---|---|
| Growth stays human | Elephant Agent should preserve the user's judgment and learning while agents do more work. | Why Elephant Agent |
| Personal Model first | Elephant Agent keeps an explicit, inspectable model of what it understands, rather than treating every retrieved memory as truth. | Personal Model first |
| Paths after understanding | Mother uses the Personal Model to shape living Paths across work and life. | Paths, Steps, and Herds |
| Curious by design | It does not wait for you to explain everything forever. It may ask when a missing or stale answer would change future help. | Proactive curiosity |
| Correctable understanding | Claims can be remembered, corrected, forgotten, disputed, and traced back to evidence. | Correctable understanding |
| Continuity across surfaces | macOS app, CLI, Dashboard, messaging, jobs, skills, tools, and recall all orbit the same local understanding system. | Continuity |
The docs map
How Elephant Agent is organized
| Area | Product question | Main docs |
|---|---|---|
| Personal Model | What does Elephant Agent understand about the person? | Personal Model first, Memory |
| Paths | How does understanding become a direction the user can keep moving? | Paths, Steps, and Herds, Why Elephant Agent |
| Memory architecture | What becomes durable truth, what stays evidence, and what is recalled only for the current turn? | System model, Memory, Embeddings |
| Daily surfaces | Where do I talk to it, inspect it, and correct it? | macOS app, CLI / Chat TUI, Dashboard |
| Visible capabilities | What can it reach for without becoming a feature shelf? | Skills, Tools, Messaging |
| Learning loops | How does understanding deepen over time? | Proactive curiosity, Background learning, Correctable understanding |
Design source of truth
These docs are the public operator guide. The deeper repository architecture truth lives in the system-design docs and the paper:
- System layer model
- Paper for the outward-facing technical report