NextClaw Roadmap

This document describes NextClaw's current priorities and execution plan. All work serves the four core pillars defined in the Product Vision. Iteration history: docs/logs.

---

Current Stage Priorities

1. Self-Awareness (The Agent Knows Itself)

Serves: Self-Aware & Self-Governing

• The agent can query its own version, configuration, runtime state, health, capability boundaries, and documentation.
• Covers static information (config, version) and dynamic information (runtime state, session context).
• Answers questions like "What channels do I have?", "What model am I using?", "Am I healthy?" in conversation.

2. Self-Governance (The Agent Manages Itself)

Serves: Self-Aware & Self-Governing

• The agent can perform management actions via tools: modify config, enable/disable channels, add/remove cron jobs, install/uninstall plugins, trigger diagnostics/restarts, etc.
• Destructive operations require confirmation.
• Users accomplish through natural language what previously required UI or CLI.

3. Plugin SDK and Development Workflow

Serves: Plugin-Based Extensibility & Ecosystem Growth

• Stabilize the plugin SDK and OpenClaw-compatible development workflow.
• Define clear boundaries between core and plugins; converge core APIs.
• Support install, enable, configure, and uninstall plugins from the UI.

4. Out-of-the-Box Experience

Serves: Out-of-the-Box Experience

• Showcase core capabilities and typical scenarios on first launch.
• Pre-built scenario templates with one-click activation.
• Optimize the path from installation to first working scenario.

---

Next Stage Directions

Marketplace Ecosystem

Serves: Plugin-Based Extensibility & Ecosystem Growth

• Plugin/skill publishing and discovery workflow.
• Infrastructure for ratings, version management, and dependency resolution.
• Close the loop: use → build → share → discover.

Host System and Data Capabilities

Serves: Infrastructure for Digital Omnipotence

• Expand tool system: file management, process management, system monitoring, and other host operations.
• Internet data access: search, scraping, API calls, information aggregation.
• User data integration: connect data scattered across apps and platforms.

Multi-Agent and Multi-Instance

• Multi-agent: session isolation, bindings, routing and runtime semantics.
• Multi-instance: deployment model, config boundaries, and best practices.

---

Ongoing Focus

• Landing page: Product value proposition and key entry point optimization.
• UI and interaction: Config UI, information architecture, consistent experience.
• OpenClaw compatibility: Track plugin SDK and channel protocols.
• Cron / Heartbeat: Harden automation and docs.
• Observability and ops: doctor, status, config hot-reload, deployment docs.
• Documentation and tutorials: Complete path for install, config, first bot, first cron.

---

How to Contribute

• Feature ideas and bugs: GitHub Issues.
• Releases and changes: per-package CHANGELOG.md and docs/logs.