wolli

Wolli

Create agents that grow around a purpose.

Install · How it works · Roadmap

What is my purpose?

Wolli lets you create agents that grow around a purpose. Each agent remembers across sessions, runs on schedules, reacts to events, and extends itself over time by writing the skills, integrations, and workflows it needs to do its job better.

Install

npm install -g wolli
wolli

The first run sets up your provider and creates your first agent. A new agent opens by interviewing you to work out its purpose, then writes its own SOUL.md — its first line becomes the agent's description everywhere. Agents and state live under ~/.wolli.

 Agents → inbox Triage my email each morning, draft replies to the routine ones, flag what needs me. scout Watch the repos and deps we ship; when a release or CVE needs action, open an issue and ping me. ledger Track project spend across providers, reconcile invoices weekly, warn me before a budget tips over. ↑/↓ browse · enter chat · tab details · type to search commands · ctrl+c quit

How it works

• Purpose-built. The agent works its purpose out with you in its first conversation and records it as the first line of its SOUL.md. It decides what the agent stores, when it speaks up, and what it does unattended.

• Self-extending. The agent builds itself out for its purpose. It curates its own memory and authors and installs its own skills, tools, workflows, and integrations; they live in its home and load on reload. The agent grows more capable at its job instead of staying a fixed tool.

• Persistent. Sessions are an append-only JSONL tree, the agent's lifetime memory. Nothing is rewritten; context is reconstructed deterministically and the latest leaf resumes by default.

• Curated memory. Three files are read once and frozen into the system prompt per session, and the agent maintains them through a memory tool:

  File	Holds
  SOUL.md	Identity, authored by the agent; the first line is its purpose.
  MEMORY.md	Durable notes the agent keeps.
  USER.md	Facts about its human.

• Always on, locally. A per-agent daemon supervised by launchd (macOS) or systemd (Linux) runs the agent on schedules and events while your machine is on.

• Sandboxed. The agent runs in a sandbox by default: srt (Apple Seatbelt / bubblewrap), or optional Docker. Reaching your real machine is an explicit, approval-gated escalation.

• Any model. Multi-provider via OAuth /login (Anthropic, OpenAI, and others).

Lifecycle

An agent is live from the moment it is created: its daemon is installed as an OS service at creation and runs on schedules and events from birth. While its SOUL.md is empty it interviews you to work out its purpose, then authors SOUL.md itself.

wolli delete <name> removes an agent and its state.

Extending an agent

An agent's capabilities are plugins under its own home, so deleting the agent removes them with it:

Two integrations ship bundled: Telegram (bidirectional chat) and a scheduler (cron). Manage plugins with wolli <agent> plugins ....

CLI

Documentation

Full documentation is in packages/wolli/docs: workflows, integrations, tools, providers, hooks, skills, prompt templates, themes, plugins, and the SDK.

Roadmap

Cloud sync, hosted email identity, agent-to-agent messaging, durable runtimes, and more inbound integrations are not built yet. Shipped versus planned is tracked in ROADMAP.md.

Contributing

Wolli is a pnpm + TypeScript monorepo (Node >= 22.19).

pnpm install
pnpm build
pnpm test

See AGENTS.md for development rules before opening a pull request.