Proactivity (Proactive Agent) for Openclaw

An AI agent skill that enables autonomous anticipation of user needs, self-healing workflows, and persistent momentum during complex tasks.

ivangdavila
v1.0.1
Mar 10, 2026
38
28.8k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install proactivity

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install proactivity using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Proactivity (Proactive Agent)?

Proactivity transforms your AI from a passive prompt-follower into a high-functioning operator that thinks ahead and acts like a true partner. By leveraging a localized memory structure, this skill allows agents to identify missing steps, surface hidden blockers, and provide concrete progress packets before the user even asks. It is an essential component for users seeking high-performance Openclaw Skills, ensuring that the agent remains resourceful and resilient across long-running tasks.

The skill operates on the principle of reverse prompting, where the agent surfaces ideas, drafts, and next steps that provide immediate value. This reduces the cognitive load on the user by keeping work moving through interruptions and context compaction. With built-in self-healing and recovery flows, the Proactivity skill ensures that your agent is always one step ahead, making it one of the most powerful Openclaw Skills for professional developers and power users.

Proactivity (Proactive Agent) Use Cases

  • Maintaining momentum on complex coding or research tasks without waiting for user input.
  • Recovering task context instantly after interruptions or context window compaction.
  • Surfacing draft fixes and prepared options proactively to reduce user overhead.
  • Following through on stale blockers or deadlines using the integrated heartbeat system.
  • Self-healing local process issues before escalating to the user for assistance.

How Proactivity (Proactive Agent) Works

  1. The skill initializes a local state directory in ~/proactivity/ to store stable activation rules and session breadcrumbs.
  2. It continuously monitors the current session state to identify the objective, last decision, and the logical next move.
  3. Through reverse prompting, the agent proposes concrete drafts or actions instead of asking open-ended questions, ensuring constant progress.
  4. During long tasks, it utilizes a working buffer to ensure fast context recovery if the session is interrupted or compacted.
  5. It executes recurring heartbeat checks to follow up on promises or deadlines within predefined safety boundaries and learned patterns.

Proactivity (Proactive Agent) Setup

To get started with this skill, install it via the hub and run the setup script to initialize the local memory structure required for Openclaw Skills behavior.

clawhub install proactivity
# Follow the setup guide to initialize the memory directory
cat ~/proactivity/setup.md

Proactivity (Proactive Agent) Data Schema & Taxonomy

The skill organizes its internal logic and state within the ~/proactivity/ directory using the following schema:

File Purpose
memory.md Durable boundary rules and stable activation preferences.
session-state.md Tracks the current objective, blockers, and the next planned move.
heartbeat.md Stores lightweight recurring check items and follow-up tasks.
patterns.md A repository of successful proactive moves and reusable execution logic.
log.md A chronological record of recent proactive actions and their outcomes.
working-buffer.md Volatile breadcrumbs used for context recovery during long or complex tasks.

Proactivity (Proactive Agent) Advanced Features

  • Self-healing capabilities to diagnose, adapt, and retry failed workflows automatically before requesting help.
  • Reverse prompting logic designed to provide high-value suggestions while avoiding user alert fatigue.
  • Workspace integration for AGENTS, TOOLS, and SOUL files to provide deep environmental awareness for Openclaw Skills.
  • Domain-specific overrides that allow for customized proactive behavior based on the specific project type.
  • Integration with the heartbeat system to manage long-term follow-ups and deadlines without manual prompts.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*