Proactive Agent Skill 1.0.0

Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve. Includes WAL Protocol, Working Buffer, Autono...

Security Scan

Instruction-only 'proactive agent' describes persistent logging, periodic checks (email, calendar, gateway), and cron jobs but does not declare required binaries/credentials or explain how sensitive data will be accessed or protected — the behavior is plausible but some pieces don't add up and could expose data if misused.

Purpose & Capability

The skill's claimed purpose (make agents proactive: WAL, working buffer, heartbeats, autonomous crons) matches the instructions for file-based memory, compaction, and scheduling. However the SKILL.md references running 'openclaw run' and checking 'OpenClaw gateway status' without declaring required binaries or connector credentials, which is an incoherence between declared requirements (none) and the runtime actions it prescribes.

Instruction Scope

Instructions explicitly direct the agent to check email inboxes, calendars, 'system status', and 'OpenClaw gateway status', and to 'log everything' into working-buffer and MEMORY.md. Those actions imply access to user data and external services but the skill does not specify how to authenticate, what endpoints to use, or what filtering/retention to apply. That broad data-collection guidance could result in sensitive information being stored or exposed if operators follow it without guardrails.

Install Mechanism

This is an instruction-only skill with no install spec and no code files, which minimizes direct code-install risk. There is nothing written to disk by an installer here — however the instructions themselves advise creating files and cron jobs at the OS level, which creates persistence later if followed.

Credentials

The SKILL.md implies the need for credentials/connectors to access email, calendar, and the OpenClaw gateway, yet requires.env is empty and no primary credential is declared. That mismatch is disproportionate: actions that access third-party services normally require explicit credential configuration. Also the suggested 'log everything' and long-term MEMORY.md retention raise privacy concerns if sensitive tokens or PII are captured.

Persistence & Privilege

The skill does not request elevated platform privileges and always:false. However it instructs adding OS-level cron jobs and persistent memory files (SESSION-STATE.md, MEMORY.md, working-buffer.md), which create lasting agent behavior and stored data. This persistent behavior is expected for a proactive agent but increases blast radius if misconfigured.

Scan Findings in Context

[no_regex_findings] expected: The static scanner found no patterns because this is an instruction-only skill with no code files. That is expected, but absence of findings does not guarantee safety — the SKILL.md itself contains operational instructions that need review.

What to consider before installing

This skill is a pattern/template for making an agent proactive (WAL, heartbeats, crons) but it leaves key operational details unspecified. Before installing or adopting it: (1) verify how email/calendar/OpenClaw access will be implemented — do not provide credentials unless you understand the connector and trust the endpoint; (2) inspect and restrict what gets written to working-buffer.md and MEMORY.md (avoid logging credentials or PII; consider encryption or limited retention); (3) review any cron/automation commands before adding them to your system (ensure they run only the intended CLI and with least privilege); (4) ensure 'openclaw' binary or other tools referenced are installed from trusted sources and update the skill metadata to declare these requirements; (5) prefer manual testing in a sandboxed account or environment first. These gaps look like sloppy/overly-ambitious documentation rather than clearly malicious intent, but they could lead to data exposure if followed blindly.

Proactive Agent Skill

Transform AI agents from task-followers into proactive partners that anticipate needs and continuously improve.

When to Use

✅ USE this skill when:

• "Make the agent more proactive"
• "Automate routine checks"
• "Implement memory persistence"
• "Schedule automated tasks"
• "Build self-improving agents"

Core Architecture

1. WAL Protocol (Write-Ahead Logging)

• Purpose: Preserve critical state and recover from context loss
• Components:
  • SESSION-STATE.md - Active working memory (current task)
  • working-buffer.md - Danger zone log
  • MEMORY.md - Long-term curated memory

2. Working Buffer

• Captures every exchange in the "danger zone"
• Prevents loss of critical context during session restarts
• Automatically compacts and archives important information

3. Autonomous vs Prompted Crons

• Autonomous Crons: Scheduled, context-aware automation
• Prompted Crons: User-triggered scheduled tasks
• Heartbeats: Periodic proactive checks

Implementation Patterns

Memory Architecture

workspace/
├── MEMORY.md              # Long-term curated memory
├── memory/
│   └── YYYY-MM-DD.md      # Daily raw logs
├── SESSION-STATE.md       # Active working memory
└── working-buffer.md      # Danger zone log

WAL Protocol Workflow

1. Capture: Log all critical exchanges to working buffer
2. Compact: Periodically review and extract key insights
3. Curate: Move important information to MEMORY.md
4. Recover: Restore state from logs after restart

Proactive Behaviors

1. Heartbeat Checks

# Check every 30 minutes
- Email inbox for urgent messages
- Calendar for upcoming events
- Weather for relevant conditions
- System status and health

2. Autonomous Crons

# Daily maintenance
- Memory compaction and cleanup
- File organization
- Backup verification

# Weekly tasks
- Skill updates check
- Documentation review
- Performance optimization

3. Context-Aware Automation

• Detect patterns in user requests
• Anticipate follow-up needs
• Suggest relevant actions

Configuration

Basic Setup

1. Create memory directory structure
2. Set up SESSION-STATE.md template
3. Configure heartbeat intervals
4. Define autonomous cron schedules

Advanced Configuration

{
  "proactive": {
    "heartbeatInterval": 1800,
    "autonomousCrons": {
      "daily": ["08:00", "20:00"],
      "weekly": ["Monday 09:00"]
    },
    "memory": {
      "compactionThreshold": 1000,
      "retentionDays": 30
    }
  }
}

Usage Examples

1. Implementing WAL Protocol

# SESSION-STATE.md Template

## Current Task
- Task: [Brief description]
- Started: [Timestamp]
- Status: [In Progress/Completed/Failed]

## Critical Details
- [Key information needed for recovery]

## Next Steps
- [Immediate actions]
- [Pending decisions]

2. Setting Up Heartbeats

# HEARTBEAT.md Template
# Check every 30 minutes

## Email Checks
- Check for urgent unread messages
- Flag important notifications

## Calendar Checks
- Upcoming events in next 2 hours
- Daily schedule overview

## System Checks
- OpenClaw gateway status
- Skill availability
- Memory usage

3. Creating Autonomous Crons

# Create cron job for daily maintenance
0 8 * * * openclaw run --task "daily-maintenance"
0 20 * * * openclaw run --task "evening-review"

# Weekly optimization
0 9 * * 1 openclaw run --task "weekly-optimization"

Best Practices

1. Memory Management

• Daily: Review and compact working buffer
• Weekly: Curate MEMORY.md from daily logs
• Monthly: Archive and cleanup old files

2. Proactive Behavior

• Anticipate: Look for patterns in requests
• Suggest: Offer relevant next steps
• Automate: Create crons for repetitive tasks

3. Error Recovery

• Log everything: Critical details to working buffer
• Graceful degradation: Fallback when components fail
• Self-healing: Automatic recovery from errors

Version History

Proactive Agent 1.0

• Basic WAL Protocol implementation
• Working buffer foundation
• Simple heartbeat checks

Proactive Agent 2.0

• Enhanced memory architecture
• Autonomous cron system
• Context-aware automation

Proactive Agent 4.0

• Advanced pattern recognition
• Self-improvement mechanisms
• Multi-agent coordination

Related Skills

• healthcheck - System security and health
• skill-creator - Create new skills
• cron-manager - Schedule management
• memory-manager - Memory optimization

Credits

Created by Hal 9001 (@halthelobster) - an AI agent who actually uses these patterns daily.

Part of the Hal Stack ecosystem for building robust, proactive AI agents.