Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion AGENTS.md
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,10 @@ Do not invent personal facts. Retrieve them or say what is missing.
Single Node.js host process. No microservices.

- Channels are skill-installed adapters such as WhatsApp, Telegram, Slack,
Discord, and Gmail.
Discord, Gmail, Teams, and Outlook. Teams and Outlook are wired and
unit-tested but not yet proven end-to-end against live Azure Bot Framework /
Microsoft Graph — planned, not ready for production traffic without that
verification.
- Each conversation group runs in its own isolated container.
- Deus owns the runtime/session/tool/context contract.
- Claude is the default compatibility backend.
Expand Down
4 changes: 2 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ A personal AI that understands you - not just recalls things you've said. It lea

3. **Picks up where you left off** - Context carries over between sessions. Start a project Monday, come back Thursday, and it knows where you left off.

4. **Lives where you already are** - WhatsApp, Telegram, Slack, Discord, Gmail. Add only the ones you need. Your memory follows you across all of them.
4. **Lives where you already are** - WhatsApp, Telegram, Slack, Discord, Gmail, Teams, Outlook. Add only the ones you need. Your memory follows you across all of them.

5. **Private by default** - Runs on your machine in isolated containers. No cloud sync, no tracking, no data leaving your computer.

Expand Down Expand Up @@ -288,7 +288,7 @@ Select review wardens (`plan-reviewer`, `code-reviewer`, `ai-eng-warden`) can ru
|---|---|---|---|---|---|
| **Memory** | Understands you - indexes facts by meaning, recalls in context | Markdown files | Via OpenClaw | Full-text search + preference profiling | Conversation only |
| **Learning** | Adapts at the personality level - tone, judgment, suggestions | No | No | Auto-creates & refines skills | No |
| **Channels** | 5 (WhatsApp, Telegram, Slack, Discord, Gmail) | 10+ | Via OpenClaw | 15+ (WhatsApp, Telegram, Signal, Matrix...) | None |
| **Channels** | 7 (WhatsApp, Telegram, Slack, Discord, Gmail, Teams, Outlook) | 10+ | Via OpenClaw | 15+ (WhatsApp, Telegram, Signal, Matrix...) | None |
| **Isolation** | Container per conversation | Opt-in Docker | Landlock + seccomp | Per-session | None |
| **LLM support** | Claude default, OpenAI/llama.cpp opt-in | Any provider | Any (via OpenClaw) | Any (10+ providers) | Claude only |
| **Setup** | ~5 min | ~15 min | ~20 min | ~10 min | N/A |
Expand Down
Loading